Research map / Video / Video segmentation and tracking
Content-Adaptive Hierarchical Hyperprior for Neural Video Coding
Junqi Liao, Yaojun Wu, Chaoyi Lin, Zhipin Deng, Li Li, Dong Liu, Xiaoyan Sun
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 hierarchical-hyperprior-guided context fusion of previous-frame and keyframe references (ablation) | DCVC-DC with context refresh · HEVC B-E average (IP -1) | BD-rate (%): 0.0 -> -17.3 | Improved |
| add hierarchical hyperprior adaptor for quality structure (ablation) | Mb · HEVC B-E average (IP -1) | BD-rate (%): -17.3 -> -18.5 | Improved |
| add GOP reference feature as adaptor input (ablation) | Mc · HEVC B-E average (IP -1) | BD-rate (%): -18.5 -> -19.4 | Improved |
| add keyframe refresh in the HH adaptor (ablation) | Md · HEVC B-E average (IP -1) | BD-rate (%): -19.4 -> -20.0 | Improved |
| use the proposed content-adaptive NVC instead of DCVC-FM | neural video codec · Average, IP -1 | BD-rate vs VTM-23.4 LDB (%): -7.07 -> -22.58 | Improved |
| use the proposed content-adaptive NVC instead of DCVC-FM | neural video codec · Average, IP 32 | BD-rate vs VTM-23.4 LDB (%): -8.82 -> -21.02 | Improved |
| add hierarchical hyperprior codec to DCVC-DC based design | neural video codec · 1080p complexity (V100) | kMACs/pixel: 1344 -> 1530 | Hurt |
| use the proposed NVC instead of DCVC-FM (encoding time) | neural video codec · 1080p complexity (V100) | Encoding time per frame (s): 0.91 -> 1.04 | Hurt |
Closest papers by meaning
- Towards Practical Real-Time Neural Video Compression (CVPR 2025 · similarity 0.89)
- Neural Video Compression with Feature Modulation (CVPR 2024 · similarity 0.88)
- PNVC: Towards Practical INR-based Video Compression (AAAI 2025 · similarity 0.88)
- Neural Video Compression with Context Modulation (CVPR 2025 · similarity 0.88)
- Neural B-frame Video Compression with Bi-directional Reference Harmonization (NeurIPS 2025 · similarity 0.86)
- DeNC: Unleash Neural Codecs in Video Streaming with Diffusion Enhancement (AAAI 2025 · similarity 0.86)
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.
