Neural Network Assisted Depth Map Packing for Compression Using Standard Hardware Video Codecs
This addresses the need for efficient depth map compression in distributed graphics systems, offering a practical solution for resource-constrained devices, though it is incremental as it builds on existing packing and codec methods.
The paper tackles the problem of compressing depth maps for streaming by using standard hardware video codecs, which lack native support for depth data, through a packing method. It proposes a neural network-assisted variable precision packing scheme that predicts optimal precision per depth map under bitrate constraints, achieving near-optimal predictions with low overhead in integration.
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.