-
Neural Network Assisted Depth Map Packing for Compression Using Standard Hardware Video Codecs
Authors:
Matti Siekkinen,
Teemu Kämäräinen
Abstract:
Depth maps are needed by various graphics rendering and processing operations. Depth map streaming is often necessary when such operations are performed in a distributed system and it requires in most cases fast performing compression, which is why video codecs are often used. Hardware implementations of standard video codecs enable relatively high resolution and framerate combinations, even on re…
▽ More
Depth maps are needed by various graphics rendering and processing operations. Depth map streaming is often necessary when such operations are performed in a distributed system and it requires in most cases fast performing compression, which is why video codecs are often used. Hardware implementations of standard video codecs enable relatively high resolution and framerate combinations, even on resource constrained devices, but unfortunately those implementations do not currently support RGB+depth extensions. However, they can be used for depth compression by first packing the depth maps into RGB or YUV frames. We investigate depth map compression using a combination of depth map packing followed by encoding with a standard video codec. We show that the precision at which depth maps are packed has a large and nontrivial impact on the resulting error caused by the combination of the packing scheme and lossy compression when bitrate is constrained. Consequently, we propose a variable precision packing scheme assisted by a neural network model that predicts the optimal precision for each depth map given a bitrate constraint. We demonstrate that the model yields near optimal predictions and that it can be integrated into a game engine with very low overhead using modern hardware.
△ Less
Submitted 30 June, 2022;
originally announced June 2022.
-
Latency and Throughput Characterization of Convolutional Neural Networks for Mobile Computer Vision
Authors:
Jussi Hanhirova,
Teemu Kämäräinen,
Sipi Seppälä,
Matti Siekkinen,
Vesa Hirvisalo,
Antti Ylä-Jääski
Abstract:
We study performance characteristics of convolutional neural networks (CNN) for mobile computer vision systems. CNNs have proven to be a powerful and efficient approach to implement such systems. However, the system performance depends largely on the utilization of hardware accelerators, which are able to speed up the execution of the underlying mathematical operations tremendously through massive…
▽ More
We study performance characteristics of convolutional neural networks (CNN) for mobile computer vision systems. CNNs have proven to be a powerful and efficient approach to implement such systems. However, the system performance depends largely on the utilization of hardware accelerators, which are able to speed up the execution of the underlying mathematical operations tremendously through massive parallelism. Our contribution is performance characterization of multiple CNN-based models for object recognition and detection with several different hardware platforms and software frameworks, using both local (on-device) and remote (network-side server) computation. The measurements are conducted using real workloads and real processing platforms. On the platform side, we concentrate especially on TensorFlow and TensorRT. Our measurements include embedded processors found on mobile devices and high-performance processors that can be used on the network side of mobile systems. We show that there exists significant latency--throughput trade-offs but the behavior is very complex. We demonstrate and discuss several factors that affect the performance and yield this complex behavior.
△ Less
Submitted 26 March, 2018;
originally announced March 2018.
-
Dissecting the End-to-end Latency of Interactive Mobile Video Applications
Authors:
Teemu Kämäräinen,
Matti Siekkinen,
Antti Ylä-Jääski,
Wenxiao Zhang,
Pan Hui
Abstract:
In this paper we measure the step-wise latency in the pipeline of three kinds of interactive mobile video applications that are rapidly gaining popularity, namely Remote Graphics Rendering (RGR) of which we focus on mobile cloud gaming, Mobile Augmented Reality (MAR), and Mobile Virtual Reality (MVR). The applications differ from each other by the way in which the user interacts with the applicati…
▽ More
In this paper we measure the step-wise latency in the pipeline of three kinds of interactive mobile video applications that are rapidly gaining popularity, namely Remote Graphics Rendering (RGR) of which we focus on mobile cloud gaming, Mobile Augmented Reality (MAR), and Mobile Virtual Reality (MVR). The applications differ from each other by the way in which the user interacts with the application, i.e., video I/O and user controls, but they all share in common the fact that their user experience is highly sensitive to end-to-end latency. Long latency between a user control event and display update renders the application unusable. Hence, understanding the nature and origins of latency of these applications is of paramount importance. We show through extensive measurements that control input and display buffering have a substantial effect on the overall delay. Our results shed light on the latency bottlenecks and the maturity of technology for seamless user experience with these applications.
△ Less
Submitted 25 November, 2016;
originally announced November 2016.
-
A First Look at Quality of Mobile Live Streaming Experience: the Case of Periscope
Authors:
Matti Siekkinen,
Enrico Masala,
Teemu Kämäräinen
Abstract:
Live multimedia streaming from mobile devices is rapidly gaining popularity but little is known about the QoE they provide. In this paper, we examine the Periscope service. We first crawl the service in order to understand its usage patterns. Then, we study the protocols used, the typical quality of experience indicators, such as playback smoothness and latency, video quality, and the energy consu…
▽ More
Live multimedia streaming from mobile devices is rapidly gaining popularity but little is known about the QoE they provide. In this paper, we examine the Periscope service. We first crawl the service in order to understand its usage patterns. Then, we study the protocols used, the typical quality of experience indicators, such as playback smoothness and latency, video quality, and the energy consumption of the Android application.
△ Less
Submitted 3 October, 2016; v1 submitted 13 May, 2016;
originally announced May 2016.
