Research map / Video / Video segmentation and tracking
Learned Rate Control for Frame-Level Adaptive Neural Video Compression via Dynamic Neural Network
Chenhao Zhang, Wei 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 |
|---|---|---|---|
| replace R-D-lambda model with learned Rate Control Agent for rate control | Neural video codec · HEVC and UVG average | bitrate error (%): 6.05 -> 1.66 | Improved |
| use proposed dynamic-route codec instead of DCVC-HEM | Neural video codec · HEVC and UVG average | BD-Rate (%): -14.808 with BD-PSNR 0.47 | Improved |
| add Rate Control Agent to the Dynamic-Route Autoencoder (ablation) | Dynamic-Route Autoencoder · HEVC Class B | coding time: about 0.05 s per frame increase | Hurt |
| add Feature Modulation network to the autoencoder (ablation) | Dynamic-Route Autoencoder · HEVC Class B | RD performance | Improved |
| use Joint-Routes Optimization instead of basic training strategy (ablation) | Dynamic-Route Autoencoder · HEVC Class B | RD performance and bitrate range | Improved |
| use low-bitrate routes with fewer channels (ablation) | Dynamic-Route Autoencoder · 1080p frame | coding time (s): 0.742 (Route 3) -> 0.569 (Route 0) | Improved |
| compare codec maximum complexity against DCVC-HEM | Dynamic-Route Autoencoder · 1080p frame | MACs and coding time: 3.3T -> 2.8T MACs; 0.781 -> 0.742 s (Route 3) | Improved |
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 0.87)
- On Quantizing Neural Representation for Variable-Rate Video Coding (ICLR 2025 · similarity 0.86)
- Neural Video Compression with Context Modulation (CVPR 2025 · similarity 0.86)
- Task-Aware Encoder Control for Deep Video Compression (CVPR 2024 · similarity 0.85)
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.
