Research map / Video / Video segmentation and tracking
Enhancing Perceptual Quality in Video Super-Resolution through Temporally-Consistent Detail Synthesis using Diffusion Models
Claudio Rota, Marco Buzzelli, Joost van de Weijer
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 |
|---|---|---|---|
| use predicted x0 (noise-free) as temporal guidance instead of x_t (ablation) | StableVSR · REDS4 center crops | tLP: 38.16 -> 6.16 | Improved |
| add motion compensation to the temporal guidance (ablation) | StableVSR · REDS4 center crops | tLP: 18.97 -> 6.16 | Improved |
| decode latent to RGB before motion compensation (ablation) | StableVSR · REDS4 center crops | tLP: 21.17 -> 6.16 | Improved |
| add temporal conditioning instead of single-frame sampling (ablation) | StableVSR · REDS4 center crops | tLP: 14.67 -> 6.16 | Improved |
| use frame-wise sampling instead of auto-regressive sampling (ablation) | StableVSR · REDS4 center crops | tLP: 8.61 -> 6.16 | Improved |
| add backward propagation (bidirectional) to unidirectional sampling (ablation) | StableVSR · REDS4 center crops | tLP: 6.36 -> 6.16 | Improved |
| use Temporal Texture Guidance via ControlNet instead of 3D conv and temporal attention video baseline | SD x4 Upscaler · REDS4 center crops | tLP: 13.08 -> 6.16 | Improved |
| diffusion-based generation instead of regression-based VSR | StableVSR vs RVRT · REDS4 | PSNR: 32.74 -> 27.97 | Hurt |
| diffusion-based generation instead of regression-based VSR | StableVSR vs RVRT · REDS4 | LPIPS: 0.128 -> 0.097 | Improved |
| increase sampling steps from 10 to 100 (ablation) | StableVSR · REDS4 center crops | perceptual vs reconstruction | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- BasicVSR++ — BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond (CVPR 2021) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- One-Step Diffusion for Detail-Rich and Temporally Consistent Video Super-Resolution (NeurIPS 2025 · similarity 0.91)
- Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution (CVPR 2024 · similarity 0.91)
- VideoGigaGAN: Towards Detail-rich Video Super-Resolution (CVPR 2025 · similarity 0.90)
- SeeClear: Semantic Distillation Enhances Pixel Condensation for Video Super-Resolution (NeurIPS 2024 · similarity 0.90)
- PatchVSR: Breaking Video Diffusion Resolution Limits with Patch-wise Video Super-Resolution (CVPR 2025 · similarity 0.90)
- SkipDiff: Adaptive Skip Diffusion Model for High-Fidelity Perceptual Image Super-resolution (AAAI 2024 · similarity 0.89)
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.
