Research map / Video / Video segmentation and tracking
MMVC: Learned Multi-Mode Video Compression With Block-Based Prediction Mode Selection and Density-Adaptive Entropy Coding
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use block-wise FP and OFC prediction ensemble instead of FP alone (ablation) | MMVC · UVG | Bpp: 0.146 -> 0.096 | Helped |
| add OFC to FP (ablation) | MMVC · Kinetics | Bpp: 0.136 -> 0.099 | Helped |
| add FPG to the FP+OFC ensemble (ablation) | MMVC · UVG | Bpp: 0.096 -> 0.084 | Helped |
| add FPG to the FP+OFC ensemble (ablation) | MMVC · Kinetics | Bpp: 0.099 -> 0.096 | Helped |
| add Skip mode to FP+OFC+FPG (ablation) | MMVC · UVG | Bpp: 0.084 -> 0.081 | Helped |
| add Skip mode to FP+OFC+FPG (ablation) | MMVC · Kinetics | Bpp: 0.096 -> 0.088 | Helped |
| add a sparse residual path alongside dense entropy coding (ablation) | MMVC with FP+OFC+FPG · Kinetics | Additional bitrate saving: 6.5% -> 23.9% | Helped |
| add a sparse residual path alongside dense entropy coding (ablation) | MMVC with FP · Kinetics | Additional bitrate saving: 0% -> 4.1% | Helped |
| use MMVC rather than VVC low delay P | video compression · MCL-JCV | BDBR (%): VVC anchor 0; MMVC -16.73 | Helped |
| use MMVC rather than VVC low delay P | video compression · HEVC Class B | BDBR (%): VVC anchor 0; MMVC -28.28 | Helped |
| use MMVC rather than VVC low delay P | video compression · UVG | BDBR (%): MMVC 0.81; Li et al. -2.58 | Hurt |
Closest papers by meaning
- M-LVC: Multiple Frames Prediction for Learned Video Compression (CVPR 2020 · similarity 0.88)
- Coarse-To-Fine Deep Video Coding With Hyperprior-Guided Mode Prediction (CVPR 2022 · similarity 0.87)
- High Dynamic Range Video Compression: A Large-Scale Benchmark Dataset and A Learned Bit-depth Scalable Compression Algorithm (CVPR 2025 · similarity 0.85)
- Neural Video Compression with Feature Modulation (CVPR 2024 · similarity 0.84)
- FLAVC: Learned Video Compression with Feature Level Attention (CVPR 2025 · similarity 0.84)
- ECVC: Exploiting Non-Local Correlations in Multiple Frames for Contextual Video Compression (CVPR 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.