Research map / Video / Video segmentation and tracking
Content Adaptive and Error Propagation Aware Deep Video Compression
Guo Lu, Chunlei Cai, Xiaoyun Zhang, Li Chen, Wanli Ouyang, Dong Xu, Zhiyong Gao
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 error propagation aware training (T=5) (ablation) | DVC · HEVC Class C (GoP=20) | PSNR: 0.2dB | Improved |
| add error propagation aware training (ablation) | DVC · 50th reconstructed frame | PSNR: 31.50dB -> 32.11dB | Improved |
| add error propagation aware training, GoP 10 (ablation) | DVC · HEVC | BDBR: 5.49% | Improved |
| add error propagation aware training, GoP 50 (ablation) | DVC · HEVC | BDBR: 10.59% | Improved |
| increase time interval T from 2 to 5 (ablation) | DVC+EPA · HEVC | BDBR: -0.42 -> -5.59 | Improved |
| increase time interval T from 5 to 6 (ablation) | DVC+EPA · HEVC | BDBR: -5.59 -> -5.61 | No clear effect |
| add online encoder updating (ablation) | DVC · HEVC | PSNR: more than 0.5dB | Improved |
| add online encoder updating (ablation) | DVC+EPA · HEVC Class C | PSNR: ~1dB | Improved |
| add online encoder updating on one frame (ablation) | DVC+EPA · single frame example | bpp: 0.056bpp -> 0.051bpp | Improved |
| add online encoder updating on one frame (ablation) | DVC+EPA · single frame example | PSNR: 36.40dB -> 36.47dB | Improved |
| update latent features instead of encoder (LFU) (ablation) | DVC+EPA · HEVC | rate-distortion | Mixed |
| update only last encoder layers (LLU) instead of whole encoder (ablation) | DVC+EPA · HEVC | rate-distortion | Hurt |
Closest papers by meaning
- An End-to-End Learning Framework for Video Compression (TPAMI 2020 · similarity 0.87)
- Learned Video Compression (ICCV 2019 · similarity 0.86)
- Single-step Diffusion-based Video Coding with Semantic-Temporal Guidance (CVPR 2026 · similarity 0.86)
- Learning for Video Compression With Hierarchical Quality and Recurrent Enhancement (CVPR 2020 · similarity 0.86)
- Plug-and-Play Versatile Compressed Video Enhancement (CVPR 2025 · similarity 0.84)
- Few-Shot Domain Adaptation for Learned Image Compression (AAAI 2025 · similarity 0.84)
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.
