Research map / Video / Video generation
InstantViR: Real-Time Video Inverse Problem Solver with Distilled Diffusion Prior
Weimin Bai, Suzhe Xu, Yiwei Ren, Jinhua Hao, Ming Sun, Wenzheng Chen, He Sun
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 |
|---|---|---|---|
| replace original video VAE with LeanVAE via teacher-space regularized distillation (ablation) | InstantViR · Inference speed (832x480) | FPS: 13.91 -> 35.56 | Improved |
| replace original VAE with LeanVAE (ablation) | InstantViR · 50% Random Inpainting | PSNR: 30.54 -> 31.78 | Improved |
| replace original VAE with LeanVAE (ablation) | InstantViR · 4x Super-Resolution | PSNR: 34.91 -> 27.04 | Hurt |
| replace original VAE with LeanVAE (ablation) | InstantViR · Gaussian Deblur | PSNR: 31.85 -> 31.16 | Hurt |
| replace original VAE with LeanVAE (ablation) | InstantViR · Gaussian Deblur | FVD: 110.51 -> 103.45 | Improved |
| replace original VAE with LeanVAE (ablation) | InstantViR · 4x Super-Resolution | FVD: 153.13 -> 156.43 | Hurt |
| distill video diffusion prior into one-step causal solver vs SVI | InstantViR vs SVI · 50% Random Inpainting | FVD: 219.90 -> 136.06 | Improved |
| distilled solver vs VISION-XL on super-resolution | InstantViR vs VISION-XL · 4x Super-Resolution | PSNR: 35.69 -> 34.91 | Hurt |
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.
- VISION-XL — VISION-XL: High Definition Video Inverse Problem Solver using Latent Image Diffusion Models (ICCV 2025) · better on 8 of 9 shared results
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
- Solving Video Inverse Problems Using Image Diffusion Models (ICLR 2025 · similarity 0.91)
- Accelerating Video Inverse Problem Solvers with Autoregressive Diffusion Models (NeurIPS 2026 · similarity 0.90)
- Warped Diffusion: Solving Video Inverse Problems with Image Diffusion Models (NeurIPS 2024 · similarity 0.90)
- Diffusion Renderer: Neural Inverse and Forward Rendering with Video Diffusion Models (CVPR 2025 · similarity 0.89)
- InvFusion: Bridging Supervised and Zero-shot Diffusion for Inverse Problems (NeurIPS 2025 · similarity 0.89)
- One-Step Diffusion for Detail-Rich and Temporally Consistent Video Super-Resolution (NeurIPS 2025 · similarity 0.88)
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.
