Research map / Video / Video segmentation and tracking
Improving Deep Video Compression by Resolution-adaptive Flow Coding
Zhihao Hu, Zhenghao Chen, Dong Xu, Guo Lu, Wanli Ouyang, Shuhang Gu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add resolution-adaptive flow coding at frame level (RaFC-frame) (ablation) | DVC · UVG, hyperprior | PSNR at 0.055 bpp: 0.5dB improvement | Improved |
| add RaFC-block on top of RaFC-frame (ablation) | RaFC-frame · UVG, hyperprior | rate-distortion | Improved |
| replace hyperprior with auto-regressive entropy model for flow (ablation) | RaFC (Ours) · four test datasets | rate-distortion | Improved |
| use RaFC flow coding instead of DVC flow coding | DVC · HEVC Class E | motion bits at same PSNR: save up to 70% bits | Improved |
| use RaFC (hyperprior) instead of DVC | DVC · all four datasets | rate-distortion | Improved |
| larger lambda (higher bitrate) in RaFC-frame | RaFC-frame · HEVC Class E | share of high-resolution flow: 38.89% -> 63.20% | Improved |
| larger lambda (higher bitrate) in RaFC-block | RaFC-block · HEVC Class E | share of 16x16 blocks: 0.98% -> 41.91% | Improved |
| use RaFC (Ours*) instead of B-frame method AD_ICCV | AD_ICCV · VTL at 0.2 bpp | PSNR: about 0.4dB improvement | Improved |
Closest papers by meaning
- Learned Video Compression (ICCV 2019 · similarity 0.87)
- Efficient Video Compression via Content-Adaptive Super-Resolution (ICCV 2021 · similarity 0.86)
- Offline and Online Optical Flow Enhancement for Deep Video Compression (AAAI 2024 · similarity 0.85)
- DPFlow: Adaptive Optical Flow Estimation with a Dual-Pyramid Framework (CVPR 2025 · similarity 0.84)
- FLAVC: Learned Video Compression with Feature Level Attention (CVPR 2025 · similarity 0.84)
- Towards Practical Real-Time Neural Video Compression (CVPR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
