Research map / Video / Video segmentation and tracking
Reinforced Rate Control for Neural Video Compression via Inter-Frame Rate–Distortion Awareness
Wuyang Cong, Junqi Shi, Lizhong Wang, Weijing Shi, Ming Lu, Hao Chen, Zhan Ma
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 |
|---|---|---|---|
| train the RL rate control with longer sequences (4 to 64 frames) (ablation) | RL rate control on DCVC · DCVC average | BD-Rate (%): -8.84 -> -16.90 | Improved |
| jointly decide down-sampling factor m with lambda (ablation) | RL rate control on DCVC · DCVC, lambda 2048 | PSNR (dB): 37.052 -> 37.245 | Improved |
| jointly decide down-sampling factor m with lambda (ablation) | RL rate control on DCVC · DCVC, lambda 256 | PSNR (dB): 34.138 -> 34.306 | Improved |
| use RL policy with lambda only instead of Zhang et al. under identical setup (ablation) | DCVC rate control · Zhang et al. setup, mini-GOP 4 | BD-Rate (%): -9.24 -> -11.05 on HEVC Class E | Improved |
| add RL rate control to natively hierarchical DCVC-RT | DCVC-RT · GOP 32 average | BD-Rate (%): -4.81 -> -5.50 vs Chen et al. | No clear effect |
| use RL rate control instead of Zhang et al. as the codec gets stronger | DCVC-DC · GOP 100 average | BD-Rate (%): -5.41 -> -13.93 | Improved |
| use RL rate control with down-sampling on DCVC-RT | DCVC-RT · DCVC-RT throughput | encoding FPS: 102 -> 111 | Improved |
| use RL rate control on unseen 360-degree video | DCVC · 360-degree video | rate bias: 7.6% -> 3.9% | Improved |
Closest papers by meaning
- Towards Practical Real-Time Neural Video Compression (CVPR 2025 · similarity 0.88)
- RL-RC-DoT: A Block-level RL agent for Task-Aware Video Compression (CVPR 2025 · similarity 0.88)
- Neural Video Compression with Context Modulation (CVPR 2025 · similarity 0.87)
- Neural Video Compression with Feature Modulation (CVPR 2024 · similarity 0.87)
- DeNC: Unleash Neural Codecs in Video Streaming with Diffusion Enhancement (AAAI 2025 · similarity 0.87)
- Combining Frame and GOP Embeddings for Neural Video Representation (CVPR 2024 · 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.
