Research map / Video / Video segmentation and tracking
ELF-VC: Efficient Learned Flexible-Rate Video Coding
Oren Rippel, Alexander G. Anderson, Kedar Tatwawadi, Sanjay Nair, Craig Lytle, Lubomir Bourdev
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 in-loop flow predictor (ablation) | ELF-VC · UVG (PSNR) | BD-rate increase: 13% | Improved |
| add dynamic loss modulation (ablation) | ELF-VC · UVG (PSNR) | BD-rate increase: 10% | Improved |
| add state propagation across blocks and frames (ablation) | ELF-VC · UVG (PSNR) | BD-rate increase: 37% | Improved |
| use DM backbone instead of common 128 (ablation) | ELF-VC · UVG (PSNR) | BD-rate increase: 76% | Improved |
| use DM backbone instead of common 192 (ablation) | ELF-VC · UVG (PSNR) | BD-rate increase: 43% | Improved |
| use asymmetric (light decoder) DM instead of symmetric 128 (ablation) | ELF-VC · UVG (PSNR) | BD-rate increase: 31% | Improved |
| use asymmetric (light decoder) DM instead of symmetric 192 (ablation) | ELF-VC · UVG (PSNR) | BD-rate increase: 18% | Improved |
| one flexible-rate range model instead of single-rate point models (ablation) | ELF-VC · UVG (PSNR) | BD-rate increase: 1% | No clear effect |
| add flow predictor (decoder speed cost) (ablation) | ELF-VC · HD 720 | encode FPS: 21 -> 19 | Hurt |
| add flexible-rate model instead of point models (parameters) (ablation) | ELF-VC · HD 720 | parameters: 220M -> 38M | Improved |
Closest papers by meaning
- Learned Video Compression (ICCV 2019 · similarity 0.91)
- Towards Practical Real-Time Neural Video Compression (CVPR 2025 · similarity 0.88)
- NVRC: Neural Video Representation Compression (NeurIPS 2024 · similarity 0.85)
- PNVC: Towards Practical INR-based Video Compression (AAAI 2025 · similarity 0.85)
- Neural Video Compression with Feature Modulation (CVPR 2024 · similarity 0.85)
- VRVVC: Variable-Rate NeRF-Based Volumetric Video Compression (AAAI 2025 · 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.
