Hierarchical B-frame Video Coding for Long Group of Pictures
This work addresses the problem of efficient video compression for random-access scenarios, offering incremental improvements over existing learned and traditional codecs.
The paper tackles the challenge of learned random-access video compression by introducing an end-to-end codec with training on long sequences and adaptive rate allocation, achieving results comparable to VVC in YUV-PSNR BD-Rate on some videos and outperforming it in VMAF BD-Rate on most test sets.
Learned video compression methods already outperform VVC in the low-delay (LD) case, but the random-access (RA) scenario remains challenging. Most works on learned RA video compression either use HEVC as an anchor or compare it to VVC in specific test conditions, using RGB-PSNR metric instead of Y-PSNR and avoiding comprehensive evaluation. Here, we present an end-to-end learned video codec for random access that combines training on long sequences of frames, rate allocation designed for hierarchical coding and content adaptation on inference. We show that under common test conditions (JVET-CTC), it achieves results comparable to VTM (VVC reference software) in terms of YUV-PSNR BD-Rate on some classes of videos, and outperforms it on almost all test sets in terms of VMAF BD-Rate. On average it surpasses open LD and RA end-to-end solutions in terms of VMAF and YUV BD-Rates.