Research map / Video / Video generation
Single-step Diffusion-based Video Coding with Semantic-Temporal Guidance
Naifu Xue, Zhaoyang Jia, Jiahao Li, Bin Li, Zihan Zheng, Yuan Zhang, Yan Lu
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 contextual semantic guidance (CSG) to the single-step diffusion generator (ablation) | S2VC · HEVC-B | DISTS BD-rate (%): 22.08 -> 14.06 (w/o CSG vs CSG only) | Improved |
| add DINOv3 semantic distillation on top of CSG (ablation) | S2VC · HEVC-B | DISTS BD-rate (%): 14.06 -> 0.00 | Improved |
| add DINOv3 semantic distillation on top of CSG (ablation) | S2VC · UVG | DISTS BD-rate (%): 29.71 -> 0.00 | Improved |
| add temporal consistency guidance (TCG) blocks (ablation) | S2VC · UVG | FloLPIPS BD-rate (%): 46.88 -> 28.46 (w/o TCG vs TCG only) | Improved |
| add cascade training (gradient propagation across frames) to TCG (ablation) | S2VC · HEVC-B | FloLPIPS BD-rate (%): 14.66 -> 0.00 | Improved |
| use S2VC instead of PLVC (perceptual codec) | video codec · HEVC-B, UVG, MCL-JCV | DISTS BD-rate saving (%): 57.69%, 32.69%, 64.49% (average 51.62%) | Improved |
| use S2VC instead of other codecs on motion-aware metric | video codec · UVG | FloLPIPS | No clear effect |
| use S2VC instead of other codecs on realism | video codec · HEVC-B, UVG, MCL-JCV | FID | Improved |
Closest papers by meaning
- One-Step Diffusion-Based Image Compression with Semantic Distillation (NeurIPS 2025 · similarity 0.91)
- From Slow Bidirectional to Fast Autoregressive Video Diffusion Models (CVPR 2025 · similarity 0.87)
- Neural Video Compression with Context Modulation (CVPR 2025 · similarity 0.87)
- Efficient Video Diffusion Models via Content-Frame Motion-Latent Decomposition (ICLR 2024 · similarity 0.87)
- MeDM: Mediating Image Diffusion Models for Video-to-Video Translation with Temporal Correspondence Guidance (AAAI 2024 · similarity 0.87)
- Neural Video Compression with Feature Modulation (CVPR 2024 · similarity 0.87)
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.
