Research map / Image generation / Super-resolution
WEVSR: Video Diffusion Generators for Real-World Video Super‑Resolution with Wavelet-Enhanced VAE Encoder
Yuying Chen, Zhirui Liu, Linyan Jiang, Qifan Gao, Xianguo Zhang, Jianhou Gan, Wenqi Ren
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 timestep sampling (ablation) | baseline · Ablation (synthetic eval) | PSNR: 22.54 -> 23.82 | Improved |
| add noise augmentation (ablation) | baseline · Ablation (synthetic eval) | PSNR: 22.54 -> 24.65 | Improved |
| add wavelet encoder (ablation) | baseline · Ablation (synthetic eval) | PSNR: 22.54 -> 25.43 | Improved |
| add wavelet encoder (ablation) | baseline · Ablation (synthetic eval) | LPIPS: 0.233 -> 0.168 | Improved |
| add timestep sampling on top of noise and wavelet (ablation) | noise + wavelet · Ablation (synthetic eval) | PSNR: 27.57 -> 28.10 | Improved |
| add timestep sampling on top of noise and wavelet (ablation) | noise + wavelet · Ablation (synthetic eval) | LPIPS: 0.143 -> 0.130 | Improved |
| add timestep sampling on top of noise and wavelet (ablation) | noise + wavelet · Ablation (synthetic eval) | SSIM: 0.839 -> 0.845 | Improved |
| add wavelet front-end to WAN VAE encoder | WAN VAE · Video VAE reconstruction | PSNR: 33.40 -> 33.65 | Improved |
| add wavelet front-end to WAN VAE encoder | WAN VAE · Video VAE reconstruction | rFVD: 3.38 -> 2.96 | Improved |
| WEVSR versus SeedVR2-7B | WEVSR · UDM10 | LPIPS: 0.203 -> 0.130 | Improved |
| WEVSR versus SeedVR2-7B | WEVSR · UDM10 | PSNR: 26.26 -> 28.10 | Improved |
| WEVSR versus SeedVR2-7B on VideoLQ | WEVSR · VideoLQ | MUSIQ: 45.76 -> 54.13 | Mixed |
Closest papers by meaning
- TurboVSR: Fantastic Video Upscalers and Where to Find Them (ICCV 2025 · similarity 0.90)
- Motion-Guided Latent Diffusion for Temporally Consistent Real-world Video Super-resolution (ECCV 2024 · similarity 0.90)
- PatchVSR: Breaking Video Diffusion Resolution Limits with Patch-wise Video Super-Resolution (CVPR 2025 · similarity 0.90)
- DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution (NeurIPS 2025 · similarity 0.88)
- Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution (CVPR 2024 · similarity 0.88)
- WF-VAE: Enhancing Video VAE by Wavelet-Driven Energy Flow for Latent Video Diffusion Model (CVPR 2025 · 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.
