Showing 51–100 of 131 results for author: Owens, A

Sound Localization by Self-Supervised Time Delay Estimation

Authors: Ziyang Chen, David F. Fouhey, Andrew Owens

Abstract: Sounds reach one microphone in a stereo pair sooner than the other, resulting in an interaural time delay that conveys their directions. Estimating a sound's time delay requires finding correspondences between the signals recorded by each microphone. We propose to learn these correspondences through self-supervision, drawing on recent techniques from visual tracking. We adapt the contrastive rando… ▽ More Sounds reach one microphone in a stereo pair sooner than the other, resulting in an interaural time delay that conveys their directions. Estimating a sound's time delay requires finding correspondences between the signals recorded by each microphone. We propose to learn these correspondences through self-supervision, drawing on recent techniques from visual tracking. We adapt the contrastive random walk of Jabri et al. to learn a cycle-consistent representation from unlabeled stereo sounds, resulting in a model that performs on par with supervised methods on "in the wild" internet recordings. We also propose a multimodal contrastive learning model that solves a visually-guided localization task: estimating the time delay for a particular person in a multi-speaker mixture, given a visual representation of their face. Project site: https://ificl.github.io/stereocrw/ △ Less

Submitted 28 January, 2023; v1 submitted 26 April, 2022; originally announced April 2022.

Comments: ECCV 2022

Cosmology Intertwined: A Review of the Particle Physics, Astrophysics, and Cosmology Associated with the Cosmological Tensions and Anomalies

Authors: Elcio Abdalla, Guillermo Franco Abellán, Amin Aboubrahim, Adriano Agnello, Ozgur Akarsu, Yashar Akrami, George Alestas, Daniel Aloni, Luca Amendola, Luis A. Anchordoqui, Richard I. Anderson, Nikki Arendse, Marika Asgari, Mario Ballardini, Vernon Barger, Spyros Basilakos, Ronaldo C. Batista, Elia S. Battistelli, Richard Battye, Micol Benetti, David Benisty, Asher Berlin, Paolo de Bernardis, Emanuele Berti, Bohdan Bidenko , et al. (178 additional authors not shown)

Abstract: In this paper we will list a few important goals that need to be addressed in the next decade, also taking into account the current discordances between the different cosmological probes, such as the disagreement in the value of the Hubble constant $H_0$, the $σ_8$--$S_8$ tension, and other less statistically significant anomalies. While these discordances can still be in part the result of system… ▽ More In this paper we will list a few important goals that need to be addressed in the next decade, also taking into account the current discordances between the different cosmological probes, such as the disagreement in the value of the Hubble constant $H_0$, the $σ_8$--$S_8$ tension, and other less statistically significant anomalies. While these discordances can still be in part the result of systematic errors, their persistence after several years of accurate analysis strongly hints at cracks in the standard cosmological scenario and the necessity for new physics or generalisations beyond the standard model. In this paper, we focus on the $5.0\,σ$ tension between the {\it Planck} CMB estimate of the Hubble constant $H_0$ and the SH0ES collaboration measurements. After showing the $H_0$ evaluations made from different teams using different methods and geometric calibrations, we list a few interesting new physics models that could alleviate this tension and discuss how the next decade's experiments will be crucial. Moreover, we focus on the tension of the {\it Planck} CMB data with weak lensing measurements and redshift surveys, about the value of the matter energy density $Ω_m$, and the amplitude or rate of the growth of structure ($σ_8,fσ_8$). We list a few interesting models proposed for alleviating this tension, and we discuss the importance of trying to fit a full array of data with a single model and not just one parameter at a time. Additionally, we present a wide range of other less discussed anomalies at a statistical significance level lower than the $H_0$--$S_8$ tensions which may also constitute hints towards new physics, and we discuss possible generic theoretical approaches that can collectively explain the non-standard nature of these signals.[Abridged] △ Less

Submitted 24 April, 2022; v1 submitted 11 March, 2022; originally announced March 2022.

Comments: Contribution to Snowmass 2021. 224 pages, 27 figures. Accepted for publication in JHEAp

Journal ref: J. High En. Astrophys. 2204, 002 (2022)

Learning Pixel Trajectories with Multiscale Contrastive Random Walks

Authors: Zhangxing Bian, Allan Jabri, Alexei A. Efros, Andrew Owens

Abstract: A range of video modeling tasks, from optical flow to multiple object tracking, share the same fundamental challenge: establishing space-time correspondence. Yet, approaches that dominate each space differ. We take a step towards bridging this gap by extending the recent contrastive random walk formulation to much denser, pixel-level space-time graphs. The main contribution is introducing hierarch… ▽ More A range of video modeling tasks, from optical flow to multiple object tracking, share the same fundamental challenge: establishing space-time correspondence. Yet, approaches that dominate each space differ. We take a step towards bridging this gap by extending the recent contrastive random walk formulation to much denser, pixel-level space-time graphs. The main contribution is introducing hierarchy into the search problem by computing the transition matrix between two frames in a coarse-to-fine manner, forming a multiscale contrastive random walk when extended in time. This establishes a unified technique for self-supervised learning of optical flow, keypoint tracking, and video object segmentation. Experiments demonstrate that, for each of these tasks, the unified model achieves performance competitive with strong self-supervised approaches specific to that task. Project webpage: https://jasonbian97.github.io/flowwalk △ Less

Submitted 4 April, 2022; v1 submitted 20 January, 2022; originally announced January 2022.

GANmouflage: 3D Object Nondetection with Texture Fields

Authors: Rui Guo, Jasmine Collins, Oscar de Lima, Andrew Owens

Abstract: We propose a method that learns to camouflage 3D objects within scenes. Given an object's shape and a distribution of viewpoints from which it will be seen, we estimate a texture that will make it difficult to detect. Successfully solving this task requires a model that can accurately reproduce textures from the scene, while simultaneously dealing with the highly conflicting constraints imposed by… ▽ More We propose a method that learns to camouflage 3D objects within scenes. Given an object's shape and a distribution of viewpoints from which it will be seen, we estimate a texture that will make it difficult to detect. Successfully solving this task requires a model that can accurately reproduce textures from the scene, while simultaneously dealing with the highly conflicting constraints imposed by each viewpoint. We address these challenges with a model based on texture fields and adversarial learning. Our model learns to camouflage a variety of object shapes from randomly sampled locations and viewpoints within the input scene, and is the first to address the problem of hiding complex object shapes. Using a human visual search study, we find that our estimated textures conceal objects significantly better than previous methods. Project site: https://rrrrrguo.github.io/ganmouflage/ △ Less

Submitted 23 April, 2023; v1 submitted 18 January, 2022; originally announced January 2022.

Current Challenges in Cepheid Distance Calibrations Using Gaia EDR3

Authors: Kayla A. Owens, Wendy L. Freedman, Barry F. Madore, Abigail J. Lee

Abstract: Using parallaxes from Gaia Early Data Release 3 (EDR3), we determine multi-wavelength BVI, JHK, and [3.6] & [4.5] micron absolute magnitudes for 37 nearby Milky Way Cepheids, covering the period range between 5 and 60 days. We apply these period-luminosity relations to Cepheids in the Large and Small Magellanic Clouds, and find that the derived distances are significantly discrepant with the geome… ▽ More Using parallaxes from Gaia Early Data Release 3 (EDR3), we determine multi-wavelength BVI, JHK, and [3.6] & [4.5] micron absolute magnitudes for 37 nearby Milky Way Cepheids, covering the period range between 5 and 60 days. We apply these period-luminosity relations to Cepheids in the Large and Small Magellanic Clouds, and find that the derived distances are significantly discrepant with the geometric distances according to detached eclipsing binaries (DEBs). We explore several potential causes of these issues, including reddening, metallicity, and the existence of an additional zero-point offset, but none provide a sufficient reconciliation with both DEB distances. We conclude that the combination of the systematic uncertainties on the EDR3 parallaxes with the uncertainties on the effect of metallicity on the Cepheid distance scale leads to a systematic error floor of approximately 3%. We therefore find that the EDR3 data is not sufficiently accurate in the regime of these bright Cepheids to determine extragalactic distances precise to the 1% level at this time, in agreement with a number of contemporary studies. △ Less

Submitted 3 January, 2022; originally announced January 2022.

Comments: 23 pages, 9 figures, 11 tables. Accepted for publication in ApJ

Structure from Silence: Learning Scene Structure from Ambient Sound

Authors: Ziyang Chen, Xixi Hu, Andrew Owens

Abstract: From whirling ceiling fans to ticking clocks, the sounds that we hear subtly vary as we move through a scene. We ask whether these ambient sounds convey information about 3D scene structure and, if so, whether they provide a useful learning signal for multimodal models. To study this, we collect a dataset of paired audio and RGB-D recordings from a variety of quiet indoor scenes. We then train mod… ▽ More From whirling ceiling fans to ticking clocks, the sounds that we hear subtly vary as we move through a scene. We ask whether these ambient sounds convey information about 3D scene structure and, if so, whether they provide a useful learning signal for multimodal models. To study this, we collect a dataset of paired audio and RGB-D recordings from a variety of quiet indoor scenes. We then train models that estimate the distance to nearby walls, given only audio as input. We also use these recordings to learn multimodal representations through self-supervision, by training a network to associate images with their corresponding sounds. These results suggest that ambient sound conveys a surprising amount of information about scene structure, and that it is a useful signal for learning multimodal features. △ Less

Submitted 10 November, 2021; originally announced November 2021.

Comments: Accepted to CoRL 2021 (Oral Presentation)

ExoMol line lists -- XLIV. IR and UV line list for silicon monoxide (SiO)

Authors: Sergei N. Yurchenko, Jonathan Tennyson, Anna-Maree Syme, Ahmad Y. Adam, Victoria H. J. Clark, Bridgette Cooper, C. Pria Dobney, Shaun T. E. Donnelly, Maire N. Gorman, Anthony E. Lynas-Gray, Thomas Meltzer, Alec Owens, Qianwei Qu, Mikhail Semenov, Wilfrid Somogyi, Apoorva Upadhyay, Samuel Wright, Juan C. Zapata Trujillo

Abstract: A new silicon monoxide ($^{28}$Si$^{16}$O) line list covering infrared, visible and ultraviolet regions called SiOUVenIR is presented. This line list extends the infrared EBJT ExoMol line list by including vibronic transitions to the $A\,{}^{1}Π$ and $E\,{}^{1}Σ^{+}$ electronic states. Strong perturbations to the $A\,{}^{1}Π$ band system are accurately modelled through the treatment of 6 dark elec… ▽ More A new silicon monoxide ($^{28}$Si$^{16}$O) line list covering infrared, visible and ultraviolet regions called SiOUVenIR is presented. This line list extends the infrared EBJT ExoMol line list by including vibronic transitions to the $A\,{}^{1}Π$ and $E\,{}^{1}Σ^{+}$ electronic states. Strong perturbations to the $A\,{}^{1}Π$ band system are accurately modelled through the treatment of 6 dark electronic states: $C\,{}^{1}Σ^{-}$, $D\,{}^{1}Δ$, $a\,{}^{3}Σ^{+}$, $b\,{}^{3}Π$, $e\,{}^{3}Σ^{-}$ and $d\,{}^{3}Δ$. Along with the $X\,{}^{1}Σ^{+}$ ground state, these 9 electronic states were used to build a comprehensive spectroscopic model of SiO using a combination of empirical and ab initio curves, including the potential energy (PE), spin-orbit (SO), electronic angular momentum (EAM) and (transition) dipole moment curves. The ab initio PE and coupling curves, computed at the multireference configuration interaction (MRCI) level of theory, were refined by fitting their analytical representations to 2617 experimentally derived SiO energy levels determined from 97 vibronic bands belonging to the $X$-$X$, $E$-$X$ and $A$-$X$ electronic systems through the MARVEL procedure. 112 observed forbidden transitions from the $C$-$X$, $D$-$X$, $e$-$X$, and $d$-$X$ bands were assigned using our predictions, and these could be fed back into the MARVEL procedure. The SiOUVenIR line list was computed using published ab initio transition dipole moments for the $E$-$X$ and $A$-$X$ bands; the line list is suitable for temperatures up to 10,000 K and for wavelengths longer than 140 nm. SiOUVenIR is available from www.exomol.com and the CDS database. △ Less

Submitted 8 November, 2021; originally announced November 2021.

TESS Hunt for Young and Maturing Exoplanets (THYME) VI: an 11 Myr giant planet transiting a very low-mass star in Lower Centaurus Crux

Authors: Andrew W. Mann, Mackenna L. Wood, Stephen P. Schmidt, Madyson G. Barber, James E. Owen, Benjamin M. Tofflemire, Elisabeth R. Newton, Eric E. Mamajek, Jonathan L. Bush, Gregory N. Mace, Adam L. Kraus, Pa Chia Thao, Andrew Vanderburg, Joe Llama, Christopher M. Johns-Krull, L. Prato, Asa G. Stahl, Shih-Yun Tang, Matthew J. Fields, Karen A. Collins, Kevin I. Collins, Tianjun Gan, Eric L. N. Jensen, Jacob Kamler, Richard P. Schwarz , et al. (26 additional authors not shown)

Abstract: Mature super-Earths and sub-Neptunes are predicted to be $\simeq$Jovian radius when younger than 10 Myr. Thus, we expect to find 5-15$R_\oplus$ planets around young stars even if their older counterparts harbor none. We report the discovery and validation of TOI 1227 b, a $0.85\pm0.05R_J$ (9.5$R_\oplus$) planet transiting a very low-mass star ($0.170\pm0.015M_\odot$) every 27.4 days. TOI~1227's ki… ▽ More Mature super-Earths and sub-Neptunes are predicted to be $\simeq$Jovian radius when younger than 10 Myr. Thus, we expect to find 5-15$R_\oplus$ planets around young stars even if their older counterparts harbor none. We report the discovery and validation of TOI 1227 b, a $0.85\pm0.05R_J$ (9.5$R_\oplus$) planet transiting a very low-mass star ($0.170\pm0.015M_\odot$) every 27.4 days. TOI~1227's kinematics and strong lithium absorption confirm it is a member of a previously discovered sub-group in the Lower Centaurus Crux OB association, which we designate the Musca group. We derive an age of 11$\pm$2 Myr for Musca, based on lithium, rotation, and the color-magnitude diagram of Musca members. The TESS data and ground-based follow-up show a deep (2.5\%) transit. We use multiwavelength transit observations and radial velocities from the IGRINS spectrograph to validate the signal as planetary in nature, and we obtain an upper limit on the planet mass of $\simeq0.5 M_J$. Because such large planets are exceptionally rare around mature low-mass stars, we suggest that TOI 1227 b is still contracting and will eventually turn into one of the more common $<5R_\oplus$ planets. △ Less

Submitted 9 March, 2022; v1 submitted 18 October, 2021; originally announced October 2021.

Comments: Accepted to the Astronomical Journal. Minor updates during referee process and proofs

Journal ref: AJ 163 156 (2022)

A Preliminary Calibration of the JAGB Method Using Gaia EDR3

Authors: Abigail J. Lee, Wendy L. Freedman, Barry F. Madore, Kayla A. Owens, In Sung Jang

Abstract: The recently-developed J-region Asymptotic Giant Branch (JAGB) method has extraordinary potential as an extragalactic standard candle, capable of calibrating the absolute magnitudes of locally-accessible Type Ia supernovae, thereby leading to an independent determination of the Hubble constant. Using Gaia Early Data Release 3 (EDR3) parallaxes, we calibrate the zeropoint of the JAGB method, based… ▽ More The recently-developed J-region Asymptotic Giant Branch (JAGB) method has extraordinary potential as an extragalactic standard candle, capable of calibrating the absolute magnitudes of locally-accessible Type Ia supernovae, thereby leading to an independent determination of the Hubble constant. Using Gaia Early Data Release 3 (EDR3) parallaxes, we calibrate the zeropoint of the JAGB method, based on the mean luminosity of a color-selected subset of carbon-rich AGB stars. We identify Galactic carbon stars from the literature and use their near-infrared photometry and Gaia EDR3 parallaxes to measure their absolute J-band magnitudes. Based on these Milky Way parallaxes we determine the zeropoint of the JAGB method to be M_J = -6.14 +/- 0.05 (stat) +/- 0.11 (sys) mag. This Galactic calibration serves as a consistency check on the JAGB zeropoint, agreeing well with previously-published, independent JAGB calibrations based on geometric, Detached-Eclipsing Binary (DEB) distances to the LMC and SMC. However, the JAGB stars used in this study suffer from the high parallax uncertainties that afflict the bright and red stars in EDR3, so we are not able to attain the higher precision of previous calibrations, and ultimately will rely on future improved DR4 and DR5 releases. △ Less

Submitted 9 October, 2021; originally announced October 2021.

Comments: 10 pages, 3 figures, 2 tables, Accepted for publication in ApJ

Theoretical rovibronic spectroscopy of the calcium monohydroxide radical (CaOH)

Authors: Alec Owens, Victoria H. J. Clark, Alexander Mitrushchenkov, Sergei N. Yurchenko, Jonathan Tennyson

Abstract: The rovibronic (rotation-vibration-electronic) spectrum of the calcium monohydroxide radical (CaOH) is of interest to studies of exoplanet atmospheres and ultracold molecules. Here, we theoretically investigate the $\tilde{A}\,^2Π$--$\tilde{X}\,^2Σ^+$ band system of CaOH using high-level \textit{ab initio} theory and variational nuclear motion calculations. New potential energy surfaces (PESs) are… ▽ More The rovibronic (rotation-vibration-electronic) spectrum of the calcium monohydroxide radical (CaOH) is of interest to studies of exoplanet atmospheres and ultracold molecules. Here, we theoretically investigate the $\tilde{A}\,^2Π$--$\tilde{X}\,^2Σ^+$ band system of CaOH using high-level \textit{ab initio} theory and variational nuclear motion calculations. New potential energy surfaces (PESs) are constructed for the $\tilde{X}\,^2Σ^+$ and $\tilde{A}\,^2Π$ electronic states along with $\tilde{A}$--$\tilde{X}$ transition dipole moment surfaces (DMSs). For the ground $\tilde{X}\,^2Σ^+$ state, a published high-level \textit{ab initio} PES is empirically refined to all available experimental rovibrational energy levels up to $J=15.5$, reproducing the observed term values with a root-mean-square (rms) error of 0.06~cm$^{-1}$. Large-scale multireference configuration interaction (MRCI) calculations using quintuple-zeta quality basis sets are employed to generate the $\tilde{A}\,^2Π$ state PESs and $\tilde{A}$--$\tilde{X}$ DMSs. Variational calculations consider both Renner-Teller and spin-orbit coupling effects, which are essential for a correct description of the spectrum of CaOH. Computed rovibronic energy levels of the $\tilde{A}\,^2Π$ state, line list calculations up to $J=125.5$, and an analysis of Renner-Teller splittings in the $ν_2$ bending mode of CaOH are discussed. △ Less

Submitted 23 July, 2021; originally announced July 2021.

Journal ref: J. Chem. Phys. 154, 234302 (2021)

A scalable helium gas cooling system for trapped-ion applications

Authors: Foni R. Lebrun-Gallagher, Nicholas Johnson, Mariam Akhtar, Sebastian Weidt, David Bretaud, Samuel J. Hile, Alexander Owens, Winfried K. Hensinger

Abstract: Microfabricated ion-trap devices offer a promising pathway towards scalable quantum computing. Research efforts have begun to focus on the engineering challenges associated with developing large-scale ion-trap arrays and networks. However, increasing the size of the array and integrating on-chip electronics can drastically increase the power dissipation within the ion-trap chips. This leads to an… ▽ More Microfabricated ion-trap devices offer a promising pathway towards scalable quantum computing. Research efforts have begun to focus on the engineering challenges associated with developing large-scale ion-trap arrays and networks. However, increasing the size of the array and integrating on-chip electronics can drastically increase the power dissipation within the ion-trap chips. This leads to an increase in the operating temperature of the ion-trap and limits the device performance. Therefore, effective thermal management is an essential consideration for any large-scale architecture. Presented here is the development of a modular cooling system designed for use with multiple ion-trapping experiments simultaneously. The system includes an extensible cryostat that permits scaling of the cooling power to meet the demands of a large network. Following experimental testing on two independent ion-trap experiments, the cooling system is expected to deliver a net cooling power of 111 W at ~70 K to up to four experiments. The cooling system is a step towards meeting the practical challenges of operating large-scale quantum computers with many qubits. △ Less

Submitted 14 June, 2021; originally announced June 2021.

Comparing Correspondences: Video Prediction with Correspondence-wise Losses

Authors: Daniel Geng, Max Hamilton, Andrew Owens

Abstract: Image prediction methods often struggle on tasks that require changing the positions of objects, such as video prediction, producing blurry images that average over the many positions that objects might occupy. In this paper, we propose a simple change to existing image similarity metrics that makes them more robust to positional errors: we match the images using optical flow, then measure the vis… ▽ More Image prediction methods often struggle on tasks that require changing the positions of objects, such as video prediction, producing blurry images that average over the many positions that objects might occupy. In this paper, we propose a simple change to existing image similarity metrics that makes them more robust to positional errors: we match the images using optical flow, then measure the visual similarity of corresponding pixels. This change leads to crisper and more perceptually accurate predictions, and does not require modifications to the image prediction network. We apply our method to a variety of video prediction tasks, where it obtains strong performance with simple network architectures, and to the closely related task of video interpolation. Code and results are available at our webpage: https://dangeng.github.io/CorrWiseLosses △ Less

Submitted 31 March, 2022; v1 submitted 19 April, 2021; originally announced April 2021.

Comments: CVPR 2022 Camera Ready

Strumming to the Beat: Audio-Conditioned Contrastive Video Textures

Authors: Medhini Narasimhan, Shiry Ginosar, Andrew Owens, Alexei A. Efros, Trevor Darrell

Abstract: We introduce a non-parametric approach for infinite video texture synthesis using a representation learned via contrastive learning. We take inspiration from Video Textures, which showed that plausible new videos could be generated from a single one by stitching its frames together in a novel yet consistent order. This classic work, however, was constrained by its use of hand-designed distance met… ▽ More We introduce a non-parametric approach for infinite video texture synthesis using a representation learned via contrastive learning. We take inspiration from Video Textures, which showed that plausible new videos could be generated from a single one by stitching its frames together in a novel yet consistent order. This classic work, however, was constrained by its use of hand-designed distance metrics, limiting its use to simple, repetitive videos. We draw on recent techniques from self-supervised learning to learn this distance metric, allowing us to compare frames in a manner that scales to more challenging dynamics, and to condition on other data, such as audio. We learn representations for video frames and frame-to-frame transition probabilities by fitting a video-specific model trained using contrastive learning. To synthesize a texture, we randomly sample frames with high transition probabilities to generate diverse temporally smooth videos with novel sequences and transitions. The model naturally extends to an audio-conditioned setting without requiring any finetuning. Our model outperforms baselines on human perceptual scores, can handle a diverse range of input videos, and can combine semantic and audio-visual cues in order to synthesize videos that synchronize well with an audio signal. △ Less

Submitted 6 April, 2021; originally announced April 2021.

Comments: Project website at https://medhini.github.io/audio_video_textures/

Planar Surface Reconstruction from Sparse Views

Authors: Linyi Jin, Shengyi Qian, Andrew Owens, David F. Fouhey

Abstract: The paper studies planar surface reconstruction of indoor scenes from two views with unknown camera poses. While prior approaches have successfully created object-centric reconstructions of many scenes, they fail to exploit other structures, such as planes, which are typically the dominant components of indoor scenes. In this paper, we reconstruct planar surfaces from multiple views, while jointly… ▽ More The paper studies planar surface reconstruction of indoor scenes from two views with unknown camera poses. While prior approaches have successfully created object-centric reconstructions of many scenes, they fail to exploit other structures, such as planes, which are typically the dominant components of indoor scenes. In this paper, we reconstruct planar surfaces from multiple views, while jointly estimating camera pose. Our experiments demonstrate that our method is able to advance the state of the art of reconstruction from sparse views, on challenging scenes from Matterport3D. Project site: https://jinlinyi.github.io/SparsePlanes/ △ Less

Submitted 20 August, 2021; v1 submitted 26 March, 2021; originally announced March 2021.

Comments: Accepted to ICCV 2021 (Oral Presentation)

ExoMol line lists -- XLI. High-temperature molecular line lists for the alkali metal hydroxides KOH and NaOH

Authors: A. Owens, J. Tennyson, S. N. Yurchenko

Abstract: Potassium hydroxide (KOH) and sodium hydroxide (NaOH) are expected to occur in the atmospheres of hot rocky super-Earth exoplanets but a lack of spectroscopic data is hampering their potential detection. Using robust first-principles methodologies, comprehensive molecular line lists for KOH and NaOH that are applicable for temperatures up to $T=3500$ K are presented. The KOH OYT4 line list covers… ▽ More Potassium hydroxide (KOH) and sodium hydroxide (NaOH) are expected to occur in the atmospheres of hot rocky super-Earth exoplanets but a lack of spectroscopic data is hampering their potential detection. Using robust first-principles methodologies, comprehensive molecular line lists for KOH and NaOH that are applicable for temperatures up to $T=3500$ K are presented. The KOH OYT4 line list covers the 0-6000 cm$^{-1}$ (wavelengths $λ> 1.67$ $μ$m) range and comprises 38 billion transitions between 7.3 million energy levels with rotational excitation up to $J=255$. The NaOH OYT5 line list covers the 0 - 9000 cm$^{-1}$ (wavelengths $λ> 1.11$ $μ$m) range and contains almost 50 billion lines involving 7.9 million molecular states with rotational excitation up to $J=206$. The OYT4 and OYT5 line lists are available from the ExoMol database at www.exomol.com and should greatly aid the study of hot rocky exoplanets. △ Less

Submitted 2 March, 2021; originally announced March 2021.

Journal ref: MNRAS, 502, 1128-1135 (2021)

The Astrophysical Distance Scale III: Distance to the Local Group Galaxy WLM using Multi-Wavelength Observations of the Tip of the Red Giant Branch, Cepheids, and JAGB Stars

Authors: Abigail J. Lee, Wendy L. Freedman, Barry F. Madore, Kayla A. Owens, Andrew J. Monson, Taylor J. Hoyt

Abstract: The local determination of the Hubble Constant sits at a crossroad. Current estimates of the local expansion rate of the Universe differ by about 1.7-sigma, derived from the Cepheid and TRGB based calibrations, applied to type Ia supernovae. To help elucidate possible sources of systematic error causing the tension, we show in this study the recently developed distance indicator, the J-region Asym… ▽ More The local determination of the Hubble Constant sits at a crossroad. Current estimates of the local expansion rate of the Universe differ by about 1.7-sigma, derived from the Cepheid and TRGB based calibrations, applied to type Ia supernovae. To help elucidate possible sources of systematic error causing the tension, we show in this study the recently developed distance indicator, the J-region Asymptotic Giant Branch (JAGB) method (arXiv:2005.10792), can serve as an independent cross-check and comparison with other local distance indicators. Furthermore, we make the case that the JAGB method has substantial potential as an independent, precise and accurate calibrator of type Ia supernovae for the determination of H0. Using the Local Group galaxy, WLM we present distance comparisons between the JAGB method, a TRGB measurement at near-infrared (JHK) wavelengths, a TRGB measurement in the optical I band, and a multi-wavelength Cepheid period-luminosity relation determination. We find: $μ_0$ (JAGB) = 24.97 +/- 0.02 (stat) +/- 0.04 (sys) mag, $μ_0$ (TRGB NIR) = 24.98 +/- 0.04 stat) +/- 0.07 (sys) mag, $μ_0$ (TRGB F814W) = 24.93 +/- 0.02 (stat) +/- 0.06 (sys) mag, $μ_0$ (Cepheids) = 24.98 +/- 0.03 (stat) +/- 0.04 (sys) mag. All four methods are in good agreement, confirming the local self-consistency of the four distance scales at the 3% level, and adding confidence that the JAGB method is as accurate and as precise a distance indicator as either of the other three astrophysically-based methods. △ Less

Submitted 8 December, 2020; originally announced December 2020.

Comments: Accepted for publication in ApJ, 21 pages, 10 figures, 5 tables

Detecting handedness of spatially oriented molecules by Coulomb explosion imaging

Authors: Cem Saribal, Alec Owens, Andrey Yachmenev, Jochen Küpper

Abstract: We present a new technique for detecting chirality in the gas phase: Chiral molecules are spatially aligned in three-dimensions by a moderately strong elliptically-polarized laser field. The momentum distributions of the charged fragments, produced by laser-induced Coulomb explosion, show distinct three-dimensional orientation of the enantiomers, when the laser polarization ellipse is rotated by a… ▽ More We present a new technique for detecting chirality in the gas phase: Chiral molecules are spatially aligned in three-dimensions by a moderately strong elliptically-polarized laser field. The momentum distributions of the charged fragments, produced by laser-induced Coulomb explosion, show distinct three-dimensional orientation of the enantiomers, when the laser polarization ellipse is rotated by a non-right angle with respect to the norm vector of the detector plane. The resulting velocity-map-image asymmetry is directly connected to the enantiomeric excess and to the absolute handedness of molecules. We demonstrated our scheme computationally for camphor (C10H16O), with its methyl-groups as marker fragments, using quantum-mechanical simulations geared toward experimentally feasible conditions. Computed sensitivity to enantiomeric excess is comparable to other modern chiroptical approaches. The present method can be readily optimized for any chiral molecule with an anisotropic polarizability tensor by adjusting the polarization state and intensity profile of the laser field. △ Less

Submitted 17 February, 2021; v1 submitted 21 September, 2020; originally announced September 2020.

Comments: 6 pages, 4 figures, accepted by JCP

MSC Class: 81-10

Self-Supervised Learning of Audio-Visual Objects from Video

Authors: Triantafyllos Afouras, Andrew Owens, Joon Son Chung, Andrew Zisserman

Abstract: Our objective is to transform a video into a set of discrete audio-visual objects using self-supervised learning. To this end, we introduce a model that uses attention to localize and group sound sources, and optical flow to aggregate information over time. We demonstrate the effectiveness of the audio-visual object embeddings that our model learns by using them for four downstream speech-oriented… ▽ More Our objective is to transform a video into a set of discrete audio-visual objects using self-supervised learning. To this end, we introduce a model that uses attention to localize and group sound sources, and optical flow to aggregate information over time. We demonstrate the effectiveness of the audio-visual object embeddings that our model learns by using them for four downstream speech-oriented tasks: (a) multi-speaker sound source separation, (b) localizing and tracking speakers, (c) correcting misaligned audio-visual data, and (d) active speaker detection. Using our representation, these tasks can be solved entirely by training on unlabeled video, without the aid of object detectors. We also demonstrate the generality of our method by applying it to non-human speakers, including cartoons and puppets.Our model significantly outperforms other self-supervised approaches, and obtains performance competitive with methods that use supervised face detection. △ Less

Submitted 10 August, 2020; originally announced August 2020.

Comments: ECCV 2020

The 2020 release of the ExoMol database: molecular line lists for exoplanet and other hot atmospheres

Authors: Jonathan Tennyson, Sergei N. Yurchenko, Ahmed F. Al-Refaie, Victoria H. J. Clark, Katy L. Chubb, Eamon K. Conway, Akhil Dewan, Maire N. Gorman, Christian Hill, A. E. Lynas-Gray, Thomas Mellor, Laura K. McKemmish, Alec Owens, Oleg L. Polyansky, Mikhail Semenov, Wilfrid Somogyi, Giovanna Tinetti, Apoorva Upadhyay, Ingo Waldmann, Yixin Wang, Samuel Wright, Olga P. Yurchenko

Abstract: The ExoMol database (www.exomol.com) provides molecular data for spectroscopic studies of hot atmospheres. While the data is intended for studies of exoplanets and other astronomical bodies, the dataset is widely applicable. The basic form of the database is extensive line lists; these are supplemented with partition functions, state lifetimes, cooling functions, Landé g-factors, temperature-depen… ▽ More The ExoMol database (www.exomol.com) provides molecular data for spectroscopic studies of hot atmospheres. While the data is intended for studies of exoplanets and other astronomical bodies, the dataset is widely applicable. The basic form of the database is extensive line lists; these are supplemented with partition functions, state lifetimes, cooling functions, Landé g-factors, temperature-dependent cross sections, opacities, pressure broadening parameters, $k$-coefficients and dipoles. This paper presents the latest release of the database which has been expanded to consider 80 molecules and 190 isotopologues totaling over 700 billion transitions. While the spectroscopic data is concentrated at infrared and visible wavelengths, ultraviolet transitions are being increasingly considered in response to requests from observers. The core of the database comes from the ExoMol project which primarily uses theoretical methods, albeit usually fine-tuned to reproduce laboratory spectra, to generate very extensive line lists for studies of hot bodies. The data has recently been supplemented by line lists deriving from direct laboratory observations, albeit usually with the use of ab initio transition intensities. A major push in the new release is towards accurate characterisation of transition frequencies for use in high resolution studies of exoplanets and other bodies. △ Less

Submitted 25 July, 2020; originally announced July 2020.

Journal ref: J. Quant. Spectrosc. Rad. Transf., 255, 107228 (2020)

Space-Time Correspondence as a Contrastive Random Walk

Authors: Allan Jabri, Andrew Owens, Alexei A. Efros

Abstract: This paper proposes a simple self-supervised approach for learning a representation for visual correspondence from raw video. We cast correspondence as prediction of links in a space-time graph constructed from video. In this graph, the nodes are patches sampled from each frame, and nodes adjacent in time can share a directed edge. We learn a representation in which pairwise similarity defines tra… ▽ More This paper proposes a simple self-supervised approach for learning a representation for visual correspondence from raw video. We cast correspondence as prediction of links in a space-time graph constructed from video. In this graph, the nodes are patches sampled from each frame, and nodes adjacent in time can share a directed edge. We learn a representation in which pairwise similarity defines transition probability of a random walk, so that long-range correspondence is computed as a walk along the graph. We optimize the representation to place high probability along paths of similarity. Targets for learning are formed without supervision, by cycle-consistency: the objective is to maximize the likelihood of returning to the initial node when walking along a graph constructed from a palindrome of frames. Thus, a single path-level constraint implicitly supervises chains of intermediate comparisons. When used as a similarity metric without adaptation, the learned representation outperforms the self-supervised state-of-the-art on label propagation tasks involving objects, semantic parts, and pose. Moreover, we demonstrate that a technique we call edge dropout, as well as self-supervised adaptation at test-time, further improve transfer for object-centric correspondence. △ Less

Submitted 3 December, 2020; v1 submitted 25 June, 2020; originally announced June 2020.

Comments: NeurIPS 2020 camera ready version -- Code at github.com/ajabri/videowalk

MARVEL analysis of the measured high-resolution rovibronic spectra of the calcium monohydroxide radical (CaOH)

Authors: Y. Wang, A. Owens, J. Tennyson, S. N. Yurchenko

Abstract: The calcium monohydroxide radical (CaOH) is an important astrophysical molecule relevant to cool stars and rocky exoplanets, amongst other astronomical environments. Here, we present a consistent set of highly accurate rovibronic (rotation-vibration-electronic) energy levels for the five lowest electronic states ($\tilde{X}\,^2Σ^+$, $\tilde{A}\,^2Π$, $\tilde{B}\,^2Σ^+$, $\tilde{C}\,^2Δ$,… ▽ More The calcium monohydroxide radical (CaOH) is an important astrophysical molecule relevant to cool stars and rocky exoplanets, amongst other astronomical environments. Here, we present a consistent set of highly accurate rovibronic (rotation-vibration-electronic) energy levels for the five lowest electronic states ($\tilde{X}\,^2Σ^+$, $\tilde{A}\,^2Π$, $\tilde{B}\,^2Σ^+$, $\tilde{C}\,^2Δ$, $\tilde{D}\,^2Σ^+$) of CaOH. A comprehensive analysis of the published spectroscopic literature on this system has allowed 1955 energy levels to be determined from 3204 rovibronic experimental transitions, all with unique quantum number labelling and measurement uncertainties. The dataset covers rotational excitation up to $J=62.5$ for molecular states below 29\,000~cm$^{-1}$. The analysis was performed using the MARVEL algorithm, which is a robust procedure based on the theory of spectroscopic networks. The dataset provided will significantly aid future interstellar, circumstellar and atmospheric detections of CaOH, as well as assisting in the design of efficient laser cooling schemes in ultracold molecule research and precision tests of fundamental physics. △ Less

Submitted 28 May, 2020; originally announced May 2020.

Journal ref: Astrophys. J. Suppl. Ser., 248, 9 (2020)

ExoMol line lists -- XXXVIII. High-temperature molecular line list of silicon dioxide (SiO$_{\mathbf{2}}$)

Authors: A. Owens, E. K. Conway, J. Tennyson, S. N. Yurchenko

Abstract: Silicon dioxide (SiO$_2$) is expected to occur in the atmospheres of hot rocky super-Earth exoplanets but a lack of spectroscopic data is hampering its possible detection. Here, we present the first, comprehensive molecular line list for SiO$_2$. The line list, named OYT3, covers the wavenumber range 0\,--\,6000~cm$^{-1}$ (wavelengths $λ> 1.67$~$μ$m) and is suitable for temperatures up to… ▽ More Silicon dioxide (SiO$_2$) is expected to occur in the atmospheres of hot rocky super-Earth exoplanets but a lack of spectroscopic data is hampering its possible detection. Here, we present the first, comprehensive molecular line list for SiO$_2$. The line list, named OYT3, covers the wavenumber range 0\,--\,6000~cm$^{-1}$ (wavelengths $λ> 1.67$~$μ$m) and is suitable for temperatures up to $T=3000$~K. Almost 33 billion transitions involving 5.69 million rotation-vibration states with rotational excitation up to $J=255$ have been computed using robust first-principles methodologies. The OYT3 line list is available from the ExoMol database at http://www.exomol.com. △ Less

Submitted 27 May, 2020; originally announced May 2020.

The high-temperature rotation-vibration spectrum and rotational clustering of silylene (SiH$_2$)

Authors: Victoria H. J. Clark, Alec Owens, Jonathan Tennyson, Sergei N. Yurchenko

Abstract: A rotation-vibration line list for the electronic ground state ($\tilde{X}^{1}A_{1}$) of SiH$_2$ is presented. The line list, named CATS, is suitable for temperatures up to 2000 K and covers the wavenumber range 0 - 10,000 cm$^{-1}$(wavelengths $>1.0$ $μ$m) for states with rotational excitation up to $J=52$. Over 310 million transitions between 593 804 energy levels have been computed variationall… ▽ More A rotation-vibration line list for the electronic ground state ($\tilde{X}^{1}A_{1}$) of SiH$_2$ is presented. The line list, named CATS, is suitable for temperatures up to 2000 K and covers the wavenumber range 0 - 10,000 cm$^{-1}$(wavelengths $>1.0$ $μ$m) for states with rotational excitation up to $J=52$. Over 310 million transitions between 593 804 energy levels have been computed variationally with a new empirically refined potential energy surface, determined by refining to 75 empirical term values with $J\leq 5$ and a newly computed high-level ab initio dipole moment surface. This is the first, comprehensive high-temperature line list to be reported for SiH$_2$ and it is expected to aid the study of silylene in plasma physics, industrial processes and possible astronomical detection. Furthermore, we investigate the phenomenon of rotational energy level clustering in the spectrum of SiH$_2$. The CATS line list is available from the ExoMol database (www.exomol.com) and the CDS database. △ Less

Submitted 9 March, 2020; originally announced April 2020.

Zodiacal Exoplanets in Time. X. The Orbit and Atmosphere of the Young "Neptune Desert"-Dwelling Planet K2-100b

Authors: E. Gaidos, T. Hirano, A. W. Mann, D. A. Owens, T. A. Berger, K. France, A. Vanderburg, H. Harakawa, K. W. Hodapp, M. Ishizuka, S. Jacobson, M. Konishi, T. Kotani, T. Kudo, T. Kurokawa, J. Nishikawa, M. Omiya, T. Serizawa, M. Tamura, A. Oeda

Abstract: We obtained high-resolution infrared spectroscopy and short-cadence photometry of the 600-800 Myr Praesepe star K2-100 during transits of its 1.67-day planet. This Neptune-size object, discovered by the NASA K2 mission, is an interloper in the "desert" of planets with similar radii on short period orbits. Our observations can be used to understand its origin and evolution by constraining the orbit… ▽ More We obtained high-resolution infrared spectroscopy and short-cadence photometry of the 600-800 Myr Praesepe star K2-100 during transits of its 1.67-day planet. This Neptune-size object, discovered by the NASA K2 mission, is an interloper in the "desert" of planets with similar radii on short period orbits. Our observations can be used to understand its origin and evolution by constraining the orbital eccentricity by transit fitting, measuring the spin-orbit obliquity by the Rossiter-McLaughlin effect, and detecting any extended, escaping hydrogen-helium envelope with the 10830A line of neutral helium in the 2s3S triplet state. Transit photometry with 1-min cadence was obtained by the K2 satellite during Campaign 18 and transit spectra were obtained with the IRD spectrograph on the Subaru telescope. While the elevated activity of K2-100 prevented us from detecting the Rossiter-McLaughlin effect, the new photometry combined with revised stellar parameters allowed us to constrain the eccentricity to e < 0.15/0.28 with 90%/99% confidence. We modeled atmospheric escape as an isothermal, spherically symmetric Parker wind, with photochemistry driven by UV radiation that we estimate by combining the observed spectrum of the active Sun with calibrations from observations of K2-100 and similar young stars in the nearby Hyades cluster. Our non-detection (<5.7mA) of a transit-associated He I line limits mass loss of a solar-composition atmosphere through a T<10000K wind to <0.3Me/Gyr. Either K2-100b is an exceptional desert-dwelling planet, or its mass loss is occurring at a lower rate over a longer interval, consistent with a core accretion-powered scenario for escape. △ Less

Submitted 28 March, 2020; originally announced March 2020.

Comments: Accepted to MNRAS

CNN-generated images are surprisingly easy to spot... for now

Authors: Sheng-Yu Wang, Oliver Wang, Richard Zhang, Andrew Owens, Alexei A. Efros

Abstract: In this work we ask whether it is possible to create a "universal" detector for telling apart real images from these generated by a CNN, regardless of architecture or dataset used. To test this, we collect a dataset consisting of fake images generated by 11 different CNN-based image generator models, chosen to span the space of commonly used architectures today (ProGAN, StyleGAN, BigGAN, CycleGAN,… ▽ More In this work we ask whether it is possible to create a "universal" detector for telling apart real images from these generated by a CNN, regardless of architecture or dataset used. To test this, we collect a dataset consisting of fake images generated by 11 different CNN-based image generator models, chosen to span the space of commonly used architectures today (ProGAN, StyleGAN, BigGAN, CycleGAN, StarGAN, GauGAN, DeepFakes, cascaded refinement networks, implicit maximum likelihood estimation, second-order attention super-resolution, seeing-in-the-dark). We demonstrate that, with careful pre- and post-processing and data augmentation, a standard image classifier trained on only one specific CNN generator (ProGAN) is able to generalize surprisingly well to unseen architectures, datasets, and training methods (including the just released StyleGAN2). Our findings suggest the intriguing possibility that today's CNN-generated images share some common systematic flaws, preventing them from achieving realistic image synthesis. Code and pre-trained networks are available at https://peterwang512.github.io/CNNDetection/ . △ Less

Submitted 4 April, 2020; v1 submitted 23 December, 2019; originally announced December 2019.

Comments: Accepted to CVPR 2020

A manifestly covariant framework for causal set dynamics

Authors: Fay Dowker, Nazireen Imambaccus, Amelia Owens, Rafael Sorkin, Stav Zalel

Abstract: We propose a manifestly covariant framework for causal set dynamics. The framework is based on a structure, dubbed covtree, which is a partial order on certain sets of finite, unlabeled causal sets. We show that every infinite path in covtree corresponds to at least one infinite, unlabeled causal set. We show that transition probabilities for a classical random walk on covtree induce a classical m… ▽ More We propose a manifestly covariant framework for causal set dynamics. The framework is based on a structure, dubbed covtree, which is a partial order on certain sets of finite, unlabeled causal sets. We show that every infinite path in covtree corresponds to at least one infinite, unlabeled causal set. We show that transition probabilities for a classical random walk on covtree induce a classical measure on the $σ$-algebra generated by the stem sets. △ Less

Submitted 25 March, 2020; v1 submitted 16 October, 2019; originally announced October 2019.

Comments: 32 pages, 12 figures

Journal ref: Class. Quantum Grav. 37 (2020) 085003

Antimagic orientations of graphs with large maximum degree

Authors: Donglei Yang, Joshua Carlson, Andrew Owens, K. E. Perry, Inne Singgih, Zi-Xia Song, Fangfang Zhang, Xiaohong Zhang

Abstract: Given a digraph $D$ with $m $ arcs, a bijection $τ: A(D)\rightarrow \{1, 2, \ldots, m\}$ is an antimagic labeling of $D$ if no two vertices in $D$ have the same vertex-sum, where the vertex-sum of a vertex $u $ in $D$ under $τ$ is the sum of labels of all arcs entering $u$ minus the sum of labels of all arcs leaving $u$. We say $(D, τ)$ is an antimagic orientation of a graph $G$ if $D$ is an orien… ▽ More Given a digraph $D$ with $m $ arcs, a bijection $τ: A(D)\rightarrow \{1, 2, \ldots, m\}$ is an antimagic labeling of $D$ if no two vertices in $D$ have the same vertex-sum, where the vertex-sum of a vertex $u $ in $D$ under $τ$ is the sum of labels of all arcs entering $u$ minus the sum of labels of all arcs leaving $u$. We say $(D, τ)$ is an antimagic orientation of a graph $G$ if $D$ is an orientation of $G$ and $τ$ is an antimagic labeling of $D$. Motivated by the conjecture of Hartsfield and Ringel from 1990 on antimagic labelings of graphs, Hefetz, Mütze, and Schwartz in 2010 initiated the study of antimagic orientations of graphs, and conjectured that every connected graph admits an antimagic orientation. This conjecture seems hard, and few related results are known. However, it has been verified to be true for regular graphs and biregular bipartite graphs. In this paper, we prove that every connected graph $G$ on $n\ge9$ vertices with maximum degree at least $n-5$ admits an antimagic orientation. △ Less

Submitted 16 August, 2019; originally announced August 2019.

Comments: 11 pages, 2 figures

Detecting Photoshopped Faces by Scripting Photoshop

Authors: Sheng-Yu Wang, Oliver Wang, Andrew Owens, Richard Zhang, Alexei A. Efros

Abstract: Most malicious photo manipulations are created using standard image editing tools, such as Adobe Photoshop. We present a method for detecting one very popular Photoshop manipulation -- image warping applied to human faces -- using a model trained entirely using fake images that were automatically generated by scripting Photoshop itself. We show that our model outperforms humans at the task of reco… ▽ More Most malicious photo manipulations are created using standard image editing tools, such as Adobe Photoshop. We present a method for detecting one very popular Photoshop manipulation -- image warping applied to human faces -- using a model trained entirely using fake images that were automatically generated by scripting Photoshop itself. We show that our model outperforms humans at the task of recognizing manipulated images, can predict the specific location of edits, and in some cases can be used to "undo" a manipulation to reconstruct the original, unedited image. We demonstrate that the system can be successfully applied to real, artist-created image manipulations. △ Less

Submitted 5 September, 2019; v1 submitted 13 June, 2019; originally announced June 2019.

Learning Individual Styles of Conversational Gesture

Authors: Shiry Ginosar, Amir Bar, Gefen Kohavi, Caroline Chan, Andrew Owens, Jitendra Malik

Abstract: Human speech is often accompanied by hand and arm gestures. Given audio speech input, we generate plausible gestures to go along with the sound. Specifically, we perform cross-modal translation from "in-the-wild'' monologue speech of a single speaker to their hand and arm motion. We train on unlabeled videos for which we only have noisy pseudo ground truth from an automatic pose detection system.… ▽ More Human speech is often accompanied by hand and arm gestures. Given audio speech input, we generate plausible gestures to go along with the sound. Specifically, we perform cross-modal translation from "in-the-wild'' monologue speech of a single speaker to their hand and arm motion. We train on unlabeled videos for which we only have noisy pseudo ground truth from an automatic pose detection system. Our proposed model significantly outperforms baseline methods in a quantitative comparison. To support research toward obtaining a computational understanding of the relationship between gesture and speech, we release a large video dataset of person-specific gestures. The project website with video, code and data can be found at http://people.eecs.berkeley.edu/~shiry/speech2gesture . △ Less

Submitted 10 June, 2019; originally announced June 2019.

Comments: CVPR 2019

Field-induced diastereomers for chiral separation

Authors: Andrey Yachmenev, Jolijn Onvlee, Emil Zak, Alec Owens, Jochen Küpper

Abstract: A novel approach for the state-specific enantiomeric enrichment and the spatial separation of enantiomers is presented. Our scheme utilizes techniques from strong-field laser physics, specifically an optical centrifuge in conjunction with a static electric field, to create a chiral field with defined handedness. Molecular enantiomers experience unique rotational excitation dynamics and this can be… ▽ More A novel approach for the state-specific enantiomeric enrichment and the spatial separation of enantiomers is presented. Our scheme utilizes techniques from strong-field laser physics, specifically an optical centrifuge in conjunction with a static electric field, to create a chiral field with defined handedness. Molecular enantiomers experience unique rotational excitation dynamics and this can be exploited to spatially separate the enantiomers using electrostatic deflection. Notably, the rotational-state-specific enantiomeric enhancement and its handedness is fully controllable. To explain these effects, we introduce the conceptual framework of $field\text{-}induced~diastereomers$ of a chiral molecule and perform robust quantum mechanical simulations on the prototypical chiral molecule propylene oxide (C$_3$H$_6$O), for which ensembles with an enantiomeric excess of up to $30~\%$ were obtained. △ Less

Submitted 17 May, 2019; originally announced May 2019.

Journal ref: Phys. Rev. Lett. 123, 243202 (2019)

Universal behaviour of diatomic halo states and the mass sensitivity of their properties

Authors: Alec Owens, Vladimir Špirko

Abstract: The scattering and spectroscopic properties of molecular halo states can serve as sensitive probes of the constancy of the electron-to-proton mass ratio $β=m_e/m_p$. Since halo states are formed by resonant $s$-wave interactions, their properties exhibit universal correlations that are fairly independent of the interactions at short distances. For diatomic molecules, these properties depend on a s… ▽ More The scattering and spectroscopic properties of molecular halo states can serve as sensitive probes of the constancy of the electron-to-proton mass ratio $β=m_e/m_p$. Since halo states are formed by resonant $s$-wave interactions, their properties exhibit universal correlations that are fairly independent of the interactions at short distances. For diatomic molecules, these properties depend on a single parameter only, and so this `universality' means that all the characteristics of a diatomic halo state can be determined with high precision if only one parameter is accurately known. Furthermore, this knowledge can be used to establish the respective property mass sensitivities for investigating the stability of $β$. Here, we show for the halo states of the helium dimers that the relationship between the probed properties and their mass sensitivity can be derived from numerically exact solutions of suitable radial Schrödinger equations for a set of effective potential energy curves. The resulting relations exhibit a weak dependence on the short-range part of the used potentials and a near-negligible dependence on the `higher-order' nonadiabatic, relativistic, QED and residual retardation effects. The presented approach is thus a robust alternative to other literature approaches, particularly in cases where a lack of experimental data prevents an accurate interaction potential from being determined. △ Less

Submitted 13 May, 2019; originally announced May 2019.

Journal ref: J. Phys. B: At. Mol. Opt. Phys. {\bf 52}, 025102 (2019)

Theoretical rotation-vibration spectroscopy of {\it cis}- and {\it trans}-diphosphene (P$_2$H$_2$) and the deuterated species P$_2$HD

Authors: Alec Owens, Sergei N. Yurchenko

Abstract: Growing astronomical interest in phosphorous (P) chemistry is stimulating the search for new interstellar P-bearing molecules; a task requiring detailed knowledge of the microwave and infrared molecular spectrum. In this work, we present comprehensive rotation-vibration line lists of the \cis- and \trans-isomers of diphosphene (P$_2$H$_2$). The line lists have been generated using robust, first-pr… ▽ More Growing astronomical interest in phosphorous (P) chemistry is stimulating the search for new interstellar P-bearing molecules; a task requiring detailed knowledge of the microwave and infrared molecular spectrum. In this work, we present comprehensive rotation-vibration line lists of the \cis- and \trans-isomers of diphosphene (P$_2$H$_2$). The line lists have been generated using robust, first-principles methodologies based on newly computed, high-level \ai\ potential energy and dipole moment surfaces. Transitions are considered between states with energies up to $8000$~cm$^{-1}$ and total angular momentum $J\leq25$. These are the first-ever line lists to be reported for P$_2$H$_2$ and they should significantly facilitate future spectroscopic characterization of this system. The deuterated species \trans-P$_2$HD and the effect of its dynamic dipole moment on the rovibrational spectrum is also discussed. △ Less

Submitted 13 May, 2019; originally announced May 2019.

Hyperfine-resolved rotation-vibration line list of ammonia (NH$_3$)

Authors: Phillip A. Coles, Alec Owens, Jochen Küpper, Andrey Yachmenev

Abstract: A comprehensive, hyperfine-resolved rotation-vibration line list for the ammonia molecule ($^{14}$NH$_3$) is presented. The line list, which considers hyperfine nuclear quadrupole coupling effects, has been computed using robust, first principles methodologies based on a highly accurate empirically refined potential energy surface. Transitions between levels with energies below $8000$~cm$^{-1}$ an… ▽ More A comprehensive, hyperfine-resolved rotation-vibration line list for the ammonia molecule ($^{14}$NH$_3$) is presented. The line list, which considers hyperfine nuclear quadrupole coupling effects, has been computed using robust, first principles methodologies based on a highly accurate empirically refined potential energy surface. Transitions between levels with energies below $8000$~cm$^{-1}$ and total angular momentum $F\leq14$ are considered. The line list shows excellent agreement with a range of experimental data and will significantly assist future high-resolution measurements of NH$_3$, both astronomically and in the laboratory. △ Less

Submitted 18 September, 2018; originally announced September 2018.

MoSculp: Interactive Visualization of Shape and Time

Authors: Xiuming Zhang, Tali Dekel, Tianfan Xue, Andrew Owens, Qiurui He, Jiajun Wu, Stefanie Mueller, William T. Freeman

Abstract: We present a system that allows users to visualize complex human motion via 3D motion sculptures---a representation that conveys the 3D structure swept by a human body as it moves through space. Given an input video, our system computes the motion sculptures and provides a user interface for rendering it in different styles, including the options to insert the sculpture back into the original vide… ▽ More We present a system that allows users to visualize complex human motion via 3D motion sculptures---a representation that conveys the 3D structure swept by a human body as it moves through space. Given an input video, our system computes the motion sculptures and provides a user interface for rendering it in different styles, including the options to insert the sculpture back into the original video, render it in a synthetic scene or physically print it. To provide this end-to-end workflow, we introduce an algorithm that estimates that human's 3D geometry over time from a set of 2D images and develop a 3D-aware image-based rendering approach that embeds the sculpture back into the scene. By automating the process, our system takes motion sculpture creation out of the realm of professional artists, and makes it applicable to a wide range of existing video material. By providing viewers with 3D information, motion sculptures reveal space-time motion information that is difficult to perceive with the naked eye, and allow viewers to interpret how different parts of the object interact over time. We validate the effectiveness of this approach with user studies, finding that our motion sculpture visualizations are significantly more informative about motion than existing stroboscopic and space-time visualization methods. △ Less

Submitted 2 January, 2019; v1 submitted 14 September, 2018; originally announced September 2018.

Comments: UIST 2018. Project page: http://mosculp.csail.mit.edu/

Structure-based Sampling and Self-correcting Machine Learning for Accurate Calculations of Potential Energy Surfaces and Vibrational Levels

Authors: Pavlo O. Dral, Alec Owens, Sergei N. Yurchenko, Walter Thiel

Abstract: We present an efficient approach for generating highly accurate molecular potential energy surfaces (PESs) using self-correcting, kernel ridge regression (KRR) based machine learning (ML). We introduce structure-based sampling to automatically assign nuclear configurations from a pre-defined grid to the training and prediction sets, respectively. Accurate high-level \textit{ab initio} energies are… ▽ More We present an efficient approach for generating highly accurate molecular potential energy surfaces (PESs) using self-correcting, kernel ridge regression (KRR) based machine learning (ML). We introduce structure-based sampling to automatically assign nuclear configurations from a pre-defined grid to the training and prediction sets, respectively. Accurate high-level \textit{ab initio} energies are required only for the points in the training set, while the energies for the remaining points are provided by the ML model with negligible computational cost. The proposed sampling procedure is shown to be superior to random sampling and also eliminates the need for training several ML models. Self-correcting machine learning has been implemented such that each additional layer corrects errors from the previous layer. The performance of our approach is demonstrated in a case study on a published high-level \textit{ab initio} PES of methyl chloride with 44,819 points. The ML model is trained on sets of different size and then used to predict the energies for tens of thousands of nuclear configurations within seconds. The resulting datasets are utilized in variational calculations of the vibrational energy levels of CH$_3$Cl. By using both structure-based sampling and self-correction, the size of the training set can be kept small (e.g. 10\% of the points) without any significant loss of accuracy. In \textit{ab initio} rovibrational spectroscopy, it is thus possible to reduce the number of computationally costly electronic structure calculations through structure-based sampling and self-correcting KKR-based machine learning by up to 90\%. △ Less

Submitted 17 August, 2018; originally announced August 2018.

Journal ref: J. Chem. Phys. 146, 244108 (2017)

A global \textit{ab initio} dipole moment surface for methyl chloride

Authors: Alec Owens, Sergei N. Yurchenko, Andrey Yachmenev, Jonathan Tennyson, Walter Thiel

Abstract: A new dipole moment surface (DMS) for methyl chloride has been generated at the CCSD(T)/aug-cc-pVQZ(+d for Cl) level of theory. To represent the DMS, a symmetry-adapted analytic representation in terms of nine vibrational coordinates has been developed and implemented. Variational calculations of the infrared spectrum of CH$_3$Cl show good agreement with a range of experimental results. This inclu… ▽ More A new dipole moment surface (DMS) for methyl chloride has been generated at the CCSD(T)/aug-cc-pVQZ(+d for Cl) level of theory. To represent the DMS, a symmetry-adapted analytic representation in terms of nine vibrational coordinates has been developed and implemented. Variational calculations of the infrared spectrum of CH$_3$Cl show good agreement with a range of experimental results. This includes vibrational transition moments, absolute line intensities of the $ν_1$, $ν_4$, $ν_5$ and $3ν_6$ bands, and a rotation-vibration line list for both CH$_{3}{}^{35}$Cl and CH$_{3}{}^{37}$Cl including states up to $J=85$ and vibrational band origins up to $4400{\,}$cm$^{-1}$. Across the spectrum band shape and structure are well reproduced and computed absolute line intensities are comparable with highly accurate experimental measurements for certain fundamental bands. We thus recommend the DMS for future use. △ Less

Submitted 16 August, 2018; originally announced August 2018.

Journal ref: J. Quant. Spectrosc. Radiat. Transf. 184, 100 (2016)

Anomalous phosphine sensitivity coefficients as probes for a possible variation of the proton-to-electron mass ratio

Authors: Alec Owens, Sergei N. Yurchenko, Vladimir Špirko

Abstract: A robust variational approach is used to investigate the sensitivity of the rotation-vibration spectrum of phosphine (PH$_3$) to a possible cosmological variation of the proton-to-electron mass ratio, $μ$. Whilst the majority of computed sensitivity coefficients, $T$, involving the low-lying vibrational states acquire the expected values of $T\approx-1$ and $T\approx-1/2$ for rotational and ro-vib… ▽ More A robust variational approach is used to investigate the sensitivity of the rotation-vibration spectrum of phosphine (PH$_3$) to a possible cosmological variation of the proton-to-electron mass ratio, $μ$. Whilst the majority of computed sensitivity coefficients, $T$, involving the low-lying vibrational states acquire the expected values of $T\approx-1$ and $T\approx-1/2$ for rotational and ro-vibrational transitions, respectively, anomalous sensitivities are uncovered for the $A_1\!-\!A_2$ splittings in the $ν_2/ν_4$, $ν_1/ν_3$ and $2ν_4^{\ell=0}/2ν_4^{\ell=2}$ manifolds of PH$_3$. A pronounced Coriolis interaction between these states in conjunction with accidentally degenerate $A_1$ and $A_2$ energy levels produces a series of enhanced sensitivity coefficients. Phosphine is expected to occur in a number of different astrophysical environments and has potential for investigating a drifting constant. Furthermore, the displayed behaviour hints at a wider trend in molecules of ${\bf C}_{3\mathrm{v}}\mathrm{(M)}$ symmetry, thus demonstrating that the splittings induced by higher-order ro-vibrational interactions are well suited for probing $μ$ in other symmetric top molecules in space, since these low-frequency transitions can be straightforwardly detected by radio telescopes. △ Less

Submitted 16 August, 2018; originally announced August 2018.

Journal ref: Mon. Not. R. Astron. Soc. 473, 4986 (2018)

Enhanced sensitivity to a possible variation of the proton-to-electron mass ratio in ammonia

Authors: Alec Owens, Sergei N. Yurchenko, Walter Thiel, Vladimir Špirko

Abstract: Numerous accidental near-degeneracies exist between the $2ν_2$ and $ν_4$ rotation-vibration energy levels of ammonia. Transitions between these two states possess significantly enhanced sensitivity to a possible variation of the proton-to-electron mass ratio $μ$. Using a robust variational approach to determine the mass sensitivity of the energy levels along with accurate experimental values for t… ▽ More Numerous accidental near-degeneracies exist between the $2ν_2$ and $ν_4$ rotation-vibration energy levels of ammonia. Transitions between these two states possess significantly enhanced sensitivity to a possible variation of the proton-to-electron mass ratio $μ$. Using a robust variational approach to determine the mass sensitivity of the energy levels along with accurate experimental values for the energies, sensitivity coefficients have been calculated for over 350 microwave, submillimetre and far infrared transitions up to $J\!=\!15$ for $^{14}$NH$_3$. The sensitivities are the largest found in ammonia to date. One particular transition, although extremely weak, has a sensitivity of $T=-16,738$ and illustrates the huge enhancement that can occur between close-lying energy levels. More promising however are a set of previously measured transitions with $T=-32$ to $28$. Given the astrophysical importance of ammonia, the sensitivities presented here confirm that $^{14}$NH$_3$ can be used exclusively to constrain a spatial or temporal variation of $μ$. Thus, certain systematic errors which affect the ammonia method can be eliminated. For all transitions analyzed we provide frequency data and Einstein A coefficients to guide future laboratory and astronomical observations. △ Less

Submitted 16 August, 2018; originally announced August 2018.

Journal ref: Phys. Rev. A 93, 052506 (2016)

A global potential energy surface and dipole moment surface for silane

Authors: Alec Owens, Sergei N. Yurchenko, Andrey Yachmenev, Walter Thiel

Abstract: A new nine-dimensional potential energy surface (PES) and dipole moment surface (DMS) for silane have been generated using high-level \textit{ab initio} theory. The PES, CBS-F12$^{\,\mathrm{HL}}$, reproduces all four fundamental term values for $^{28}$SiH$_4$ with sub-wavenumber accuracy, resulting in an overall root-mean-square (rms) error of $0.63{\,}$cm$^{-1}$. The PES is based on explicitly co… ▽ More A new nine-dimensional potential energy surface (PES) and dipole moment surface (DMS) for silane have been generated using high-level \textit{ab initio} theory. The PES, CBS-F12$^{\,\mathrm{HL}}$, reproduces all four fundamental term values for $^{28}$SiH$_4$ with sub-wavenumber accuracy, resulting in an overall root-mean-square (rms) error of $0.63{\,}$cm$^{-1}$. The PES is based on explicitly correlated coupled cluster calculations with extrapolation to the complete basis set limit, and incorporates a range of higher-level additive energy corrections to account for core-valence electron correlation, higher-order coupled cluster terms, and scalar relativistic effects. Systematic errors in computed intra-band rotational energy levels are reduced by empirically refining the equilibrium geometry. The resultant Si-H bond length is in excellent agreement with previous experimental and theoretical values. Vibrational transition moments, absolute line intensities of the $ν_3$ band, and the infrared spectrum for $^{28}$SiH$_4$ including states up to $J=20$ and vibrational band origins up to $5000{\,}$cm$^{-1}$ are calculated and compared with available experimental results. The DMS tends to marginally overestimate the strength of line intensities. Despite this, band shape and structure across the spectrum are well reproduced and show good agreement with experiment. We thus recommend the PES and DMS for future use. △ Less

Submitted 16 August, 2018; originally announced August 2018.

Journal ref: J. Chem. Phys. 143, 244317 (2015)

Accurate prediction of H$_3$O$^+$ and D$_3$O$^+$ sensitivity coefficients to probe a variable proton-to-electron mass ratio

Authors: Alec Owens, Sergei N. Yurchenko, Oleg L. Polyansky, Roman I. Ovsyannikov, Walter Thiel, Vladimir Špirko

Abstract: The mass sensitivity of the vibration-rotation-inversion transitions of H$_3{}^{16}$O$^+$, H$_3{}^{18}$O$^+$, and D$_3{}^{16}$O$^+$ is investigated variationally using the nuclear motion program TROVE~\citep{TROVE:2007}. The calculations utilize new high-level \textit{ab initio} potential energy and dipole moment surfaces. Along with the mass dependence, frequency data and Einstein A coefficients… ▽ More The mass sensitivity of the vibration-rotation-inversion transitions of H$_3{}^{16}$O$^+$, H$_3{}^{18}$O$^+$, and D$_3{}^{16}$O$^+$ is investigated variationally using the nuclear motion program TROVE~\citep{TROVE:2007}. The calculations utilize new high-level \textit{ab initio} potential energy and dipole moment surfaces. Along with the mass dependence, frequency data and Einstein A coefficients are computed for all transitions probed. Particular attention is paid to the $Δ|k|=3$ and $Δ|k-l|=3$ transitions comprising the accidentally coinciding $|J,K\!=\!0,v_2\!=\!0^+\rangle$ and $|J,K\!=\!3,v_2\!=\!0^-\rangle$ rotation-inversion energy levels. The newly computed probes exhibit sensitivities comparable to their ammonia and methanol counterparts, thus demonstrating their potential for testing the cosmological stability of the proton-to-electron mass ratio. The theoretical TROVE results are in close agreement with sensitivities obtained using the nonrigid and rigid inverter approximate models, confirming that the \textit{ab initio} theory used in the present study is adequate. △ Less

Submitted 16 August, 2018; originally announced August 2018.

Journal ref: Mon. Not. R. Astron. Soc. 454, 2292 (2015)

Accurate \textit{ab initio} vibrational energies of methyl chloride

Authors: Alec Owens, Sergei N. Yurchenko, Andrey Yachmenev, Jonathan Tennyson, Walter Thiel

Abstract: Two new nine-dimensional potential energy surfaces (PESs) have been generated using high-level \textit{ab initio} theory for the two main isotopologues of methyl chloride, CH$_{3}{}^{35}$Cl and CH$_{3}{}^{37}$Cl. The respective PESs, CBS-35$^{\,\mathrm{HL}}$ and CBS-37$^{\,\mathrm{HL}}$, are based on explicitly correlated coupled cluster calculations with extrapolation to the complete basis set (C… ▽ More Two new nine-dimensional potential energy surfaces (PESs) have been generated using high-level \textit{ab initio} theory for the two main isotopologues of methyl chloride, CH$_{3}{}^{35}$Cl and CH$_{3}{}^{37}$Cl. The respective PESs, CBS-35$^{\,\mathrm{HL}}$ and CBS-37$^{\,\mathrm{HL}}$, are based on explicitly correlated coupled cluster calculations with extrapolation to the complete basis set (CBS) limit, and incorporate a range of higher-level (HL) additive energy corrections to account for core-valence electron correlation, higher-order coupled cluster terms, scalar relativistic effects, and diagonal Born-Oppenheimer corrections. Variational calculations of the vibrational energy levels were performed using the computer program TROVE, whose functionality has been extended to handle molecules of the form XY$_3$Z. Fully converged energies were obtained by means of a complete vibrational basis set extrapolation. The CBS-35$^{\,\mathrm{HL}}$ and CBS-37$^{\,\mathrm{HL}}$ PESs reproduce the fundamental term values with root-mean-square errors of $0.75$ and $1.00{\,}$cm$^{-1}$ respectively. An analysis of the combined effect of the HL corrections and CBS extrapolation on the vibrational wavenumbers indicates that both are needed to compute accurate theoretical results for methyl chloride. We believe that it would be extremely challenging to go beyond the accuracy currently achieved for CH$_3$Cl without empirical refinement of the respective PESs. △ Less

Submitted 16 August, 2018; originally announced August 2018.

Journal ref: J. Chem. Phys. 142, 244306 (2015)

The rotation-vibration spectrum of methyl fluoride from first principles

Authors: Alec Owens, Andrey Yachmenev, Jochen Küpper, Sergei N. Yurchenko, Walter Thiel

Abstract: Accurate \textit{ab initio} calculations on the rotation-vibration spectrum of methyl fluoride (\ce{CH3F}) are reported. A new nine-dimensional potential energy surface (PES) and dipole moment surface (DMS) have been generated using high-level electronic structure methods. Notably, the PES was constructed from explicitly correlated coupled cluster calculations with extrapolation to the complete ba… ▽ More Accurate \textit{ab initio} calculations on the rotation-vibration spectrum of methyl fluoride (\ce{CH3F}) are reported. A new nine-dimensional potential energy surface (PES) and dipole moment surface (DMS) have been generated using high-level electronic structure methods. Notably, the PES was constructed from explicitly correlated coupled cluster calculations with extrapolation to the complete basis set limit and considered additional energy corrections to account for core-valence electron correlation, higher-order coupled cluster terms beyond perturbative triples, scalar relativistic effects, and the diagonal Born-Oppenheimer correction. The PES and DMS are evaluated through robust variational nuclear motion computations of pure rotational and vibrational energy levels, the equilibrium geometry of \ce{CH3F}, vibrational transition moments, absolute line intensities of the $ν_6$ band, and the rotation-vibration spectrum up to $J=40$. The computed results show excellent agreement with a range of experimental sources, in particular the six fundamentals are reproduced with a root-mean-square error of 0.69~cm$^{-1}$. This work represents the most accurate theoretical treatment of the rovibrational spectrum of \ce{CH3F} to date. △ Less

Submitted 14 August, 2018; originally announced August 2018.

Journal ref: Phys. Chem. Chem. Phys., 2018, Advance Article

Coherent Control of the Rotation Axis of Molecular Superrotors

Authors: Alec Owens, Andrey Yachmenev, Jochen Küpper

Abstract: The control of ultrafast molecular rotational motion has benefited from the development of innovative techniques in strong-field laser physics. Here, we theoretically demonstrate a novel type of coherent control by inducing rotation of an asymmetric-top molecule about two different molecular axes. An optical centrifuge is applied to the hydrogen sulfide (H$_2$S) molecule to create a molecular supe… ▽ More The control of ultrafast molecular rotational motion has benefited from the development of innovative techniques in strong-field laser physics. Here, we theoretically demonstrate a novel type of coherent control by inducing rotation of an asymmetric-top molecule about two different molecular axes. An optical centrifuge is applied to the hydrogen sulfide (H$_2$S) molecule to create a molecular superrotor, an object performing ultrafast rotation about a well-defined axis. Using two distinct pulse envelopes for the optical centrifuge, we show that H$_2$S can be excited along separate pathways of rotational states. This leads to stable rotation about two entirely different molecular axes while ensuring rotation is about the propagation direction of the centrifuge, i.e., the laboratory-fixed $Z$-axis. The presented scheme to control the angular momentum alignment of a molecule will, for instance, be useful in studies of molecule-molecule or molecule-surface scattering, especially due to the large amounts of energy associated with superrotors, which can even be controlled by changing the duration of the optical centrifuge pulse. △ Less

Submitted 11 July, 2018; originally announced July 2018.

Journal ref: J. Phys. Chem. Lett., 2018, 9 (15), pp 4206-4209

ExoMol line lists - XXIX. The rotation-vibration spectrum of methyl chloride up to 1200 K

Authors: A. Owens, A. Yachmenev, W. Thiel, A. Fateev, J. Tennyson, S. N. Yurchenko

Abstract: Comprehensive rotation-vibration line lists are presented for the two main isotopologues of methyl chloride, $^{12}$CH$_3{}^{35}$Cl and $^{12}$CH$_3{}^{37}$Cl. The line lists, OYT-35 and OYT-37, are suitable for temperatures up to $T=1200\,$K and consider transitions with rotational excitation up to $J=85$ in the wavenumber range $0$--$6400\,$cm$^{-1}$ (wavelengths $λ> 1.56\,μ$m). Over 166 billion… ▽ More Comprehensive rotation-vibration line lists are presented for the two main isotopologues of methyl chloride, $^{12}$CH$_3{}^{35}$Cl and $^{12}$CH$_3{}^{37}$Cl. The line lists, OYT-35 and OYT-37, are suitable for temperatures up to $T=1200\,$K and consider transitions with rotational excitation up to $J=85$ in the wavenumber range $0$--$6400\,$cm$^{-1}$ (wavelengths $λ> 1.56\,μ$m). Over 166 billion transitions between 10.2 million energy levels have been calculated variationally for each line list using a new empirically refined potential energy surface, determined by refining to 739 experimentally derived energy levels up to $J=5$, and an established {\it ab initio} dipole moment surface. The OYT line lists show excellent agreement with newly measured high-temperature infrared absorption cross-sections, reproducing both strong and weak intensity features across the spectrum. The line lists are available from the ExoMol database and the CDS database. △ Less

Submitted 9 June, 2018; originally announced June 2018.

Journal ref: MNRAS, 479, 3002-3010 (2018)

More Than a Feeling: Learning to Grasp and Regrasp using Vision and Touch

Authors: Roberto Calandra, Andrew Owens, Dinesh Jayaraman, Justin Lin, Wenzhen Yuan, Jitendra Malik, Edward H. Adelson, Sergey Levine

Abstract: For humans, the process of grasping an object relies heavily on rich tactile feedback. Most recent robotic grasping work, however, has been based only on visual input, and thus cannot easily benefit from feedback after initiating contact. In this paper, we investigate how a robot can learn to use tactile information to iteratively and efficiently adjust its grasp. To this end, we propose an end-to… ▽ More For humans, the process of grasping an object relies heavily on rich tactile feedback. Most recent robotic grasping work, however, has been based only on visual input, and thus cannot easily benefit from feedback after initiating contact. In this paper, we investigate how a robot can learn to use tactile information to iteratively and efficiently adjust its grasp. To this end, we propose an end-to-end action-conditional model that learns regrasping policies from raw visuo-tactile data. This model -- a deep, multimodal convolutional network -- predicts the outcome of a candidate grasp adjustment, and then executes a grasp by iteratively selecting the most promising actions. Our approach requires neither calibration of the tactile sensors, nor any analytical modeling of contact forces, thus reducing the engineering effort required to obtain efficient grasping policies. We train our model with data from about 6,450 grasping trials on a two-finger gripper equipped with GelSight high-resolution tactile sensors on each finger. Across extensive experiments, our approach outperforms a variety of baselines at (i) estimating grasp adjustment outcomes, (ii) selecting efficient grasp adjustments for quick grasping, and (iii) reducing the amount of force applied at the fingers, while maintaining competitive performance. Finally, we study the choices made by our model and show that it has successfully acquired useful and interpretable grasping behaviors. △ Less

Submitted 26 July, 2018; v1 submitted 28 May, 2018; originally announced May 2018.

Comments: 8 pages. Published on IEEE Robotics and Automation Letters (RAL). Website: https://sites.google.com/view/more-than-a-feeling

Fighting Fake News: Image Splice Detection via Learned Self-Consistency

Authors: Minyoung Huh, Andrew Liu, Andrew Owens, Alexei A. Efros

Abstract: Advances in photo editing and manipulation tools have made it significantly easier to create fake imagery. Learning to detect such manipulations, however, remains a challenging problem due to the lack of sufficient amounts of manipulated training data. In this paper, we propose a learning algorithm for detecting visual image manipulations that is trained only using a large dataset of real photogra… ▽ More Advances in photo editing and manipulation tools have made it significantly easier to create fake imagery. Learning to detect such manipulations, however, remains a challenging problem due to the lack of sufficient amounts of manipulated training data. In this paper, we propose a learning algorithm for detecting visual image manipulations that is trained only using a large dataset of real photographs. The algorithm uses the automatically recorded photo EXIF metadata as supervisory signal for training a model to determine whether an image is self-consistent -- that is, whether its content could have been produced by a single imaging pipeline. We apply this self-consistency model to the task of detecting and localizing image splices. The proposed method obtains state-of-the-art performance on several image forensics benchmarks, despite never seeing any manipulated images at training. That said, it is merely a step in the long quest for a truly general purpose visual forensics tool. △ Less

Submitted 5 September, 2018; v1 submitted 10 May, 2018; originally announced May 2018.

Audio-Visual Scene Analysis with Self-Supervised Multisensory Features

Authors: Andrew Owens, Alexei A. Efros

Abstract: The thud of a bouncing ball, the onset of speech as lips open -- when visual and audio events occur together, it suggests that there might be a common, underlying event that produced both signals. In this paper, we argue that the visual and audio components of a video signal should be modeled jointly using a fused multisensory representation. We propose to learn such a representation in a self-sup… ▽ More The thud of a bouncing ball, the onset of speech as lips open -- when visual and audio events occur together, it suggests that there might be a common, underlying event that produced both signals. In this paper, we argue that the visual and audio components of a video signal should be modeled jointly using a fused multisensory representation. We propose to learn such a representation in a self-supervised way, by training a neural network to predict whether video frames and audio are temporally aligned. We use this learned representation for three applications: (a) sound source localization, i.e. visualizing the source of sound in a video; (b) audio-visual action recognition; and (c) on/off-screen audio source separation, e.g. removing the off-screen translator's voice from a foreign official's speech. Code, models, and video results are available on our webpage: http://andrewowens.com/multisensory △ Less

Submitted 9 October, 2018; v1 submitted 10 April, 2018; originally announced April 2018.

Climbing the rotational ladder to chirality

Authors: Alec Owens, Andrey Yachmenev, Sergei N. Yurchenko, Jochen Küpper

Abstract: Molecular chirality is conventionally understood as space-inversion-symmetry breaking in the equilibrium structure of molecules. Less well known is that achiral molecules can be made chiral through extreme rotational excitation. Here, we theoretically demonstrate a clear strategy for generating rotationally-induced chirality (RIC): An optical centrifuge rotationally excites the phosphine molecule… ▽ More Molecular chirality is conventionally understood as space-inversion-symmetry breaking in the equilibrium structure of molecules. Less well known is that achiral molecules can be made chiral through extreme rotational excitation. Here, we theoretically demonstrate a clear strategy for generating rotationally-induced chirality (RIC): An optical centrifuge rotationally excites the phosphine molecule (PH$_3$) into chiral cluster states that correspond to clockwise ($R$-enantiomer) or anticlockwise ($L$-enantiomer) rotation about axes almost coinciding with single P-H bonds. Application of a strong dc electric field during the centrifuge pulse favors the production of one rotating enantiomeric form over the other, creating dynamically chiral molecules with $permanently$ oriented rotational angular momentum. This essential step toward characterizing RIC promises a fresh perspective on chirality as a fundamental aspect of nature. △ Less

Submitted 24 October, 2018; v1 submitted 21 February, 2018; originally announced February 2018.

Journal ref: Phys. Rev. Lett. 121, 193201 (2018)

RichMol: A general variational approach for rovibrational molecular dynamics in external electric fields

Authors: Alec Owens, Andrey Yachmenev

Abstract: A general variational approach for computing the rovibrational dynamics of polyatomic molecules in the presence of external electric fields is presented. Highly accurate, full-dimensional variational calculations provide a basis of field-free rovibrational states for evaluating the rovibrational matrix elements of high-rank Cartesian tensor operators, and for solving the time-dependent Schrödinger… ▽ More A general variational approach for computing the rovibrational dynamics of polyatomic molecules in the presence of external electric fields is presented. Highly accurate, full-dimensional variational calculations provide a basis of field-free rovibrational states for evaluating the rovibrational matrix elements of high-rank Cartesian tensor operators, and for solving the time-dependent Schrödinger equation. The effect of the external electric field is treated as a multipole moment expansion truncated at the second hyperpolarizability interaction term. Our fully numerical and computationally efficient method has been implemented in a new program, RichMol, which can simulate the effects of multiple external fields of arbitrary strength, polarization, pulse shape and duration. Illustrative calculations of two-color orientation and rotational excitation with an optical centrifuge of NH$_3$ are discussed. △ Less

Submitted 21 February, 2018; originally announced February 2018.

ExoMol line lists -- XXII. The rotation-vibration spectrum of silane up to 1200 K

Authors: A. Owens, A. Yachmenev, W. Thiel, J. Tennyson, S. N. Yurchenko

Abstract: A variationally computed $^{28}$SiH$_4$ rotation-vibration line list applicable for temperatures up to $T=1200\,$K is presented. The line list, called OY2T, considers transitions with rotational excitation up to $J=42$ in the wavenumber range $0$--$5000\,$cm$^{-1}$ (wavelengths $λ> 2\,μ$m). Just under 62.7 billion transitions have been calculated between 6.1 million energy levels. Rovibrational ca… ▽ More A variationally computed $^{28}$SiH$_4$ rotation-vibration line list applicable for temperatures up to $T=1200\,$K is presented. The line list, called OY2T, considers transitions with rotational excitation up to $J=42$ in the wavenumber range $0$--$5000\,$cm$^{-1}$ (wavelengths $λ> 2\,μ$m). Just under 62.7 billion transitions have been calculated between 6.1 million energy levels. Rovibrational calculations have utilized a new `spectroscopic' potential energy surface determined by empirical refinement to 1452 experimentally derived energy levels up to $J=6$, and a previously reported \textit{ab initio} dipole moment surface. The temperature-dependent partition function of silane, the OY2T line list format, and the temperature dependence of the OY2T line list are discussed. Comparisons with the PNNL spectral library and other experimental sources indicate that the OY2T line list is robust and able to accurately reproduce weaker intensity features. The full line list is available from the ExoMol database and the CDS database. △ Less

Submitted 26 December, 2017; originally announced December 2017.

Comments: 11 pages, 9 figures, 3 tables

Journal ref: Mon. Not. R. astr. Soc, 471, 5025-5032 (2017)