Research map / Video / Video segmentation and tracking
Video Dynamics Prior: An Internal Learning Approach for Robust Video Enhancements
Gaurav Shrivastava, Ser-Nam Lim, Abhinav Shrivastava
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 |
|---|---|---|---|
| Optimize VDP on the test sequence rather than use a pretrained denoiser | Video denoising · DAVIS, Poisson noise λ = 25 | PSNR: 22.08 -> 31.96 | Improved |
| Optimize VDP on the test sequence rather than use a pretrained denoiser | Video denoising · DAVIS, Poisson noise λ = 30 | PSNR: 20.98 -> 30.07 | Improved |
| Use VDP instead of RIFE for interpolation | Video frame interpolation · UCF101, clean input | PSNR: 35.41 -> 35.41 | No clear effect |
| Use VDP instead of DAIN for interpolation | Video frame interpolation · UCF101, second frame noisy, σ = 15 | PSNR: 20.87 -> 34.92 | Improved |
| Use VDP instead of BasicVSR++ for super-resolution | Video super-resolution · Vimeo-90K-T, clean input | PSNR: 35.95 -> 35.70 | Hurt |
| Use VDP instead of EDVR for super-resolution | Video super-resolution · Vimeo-90K-T, noisy input σ = 5 | PSNR: 32.02 -> 33.87 | Improved |
| Use VDP instead of FGVC for object removal | Video object removal · DAVIS, stationary mask | PSNR: 31.92 -> 32.06 | Improved |
| Add spatial pyramid loss to reconstruction loss (ablation) | VDP · DAVIS, Poisson noise λ = 25 denoising | PSNR: 20.58 -> 30.14 | Improved |
| Add variation loss to reconstruction loss (ablation) | VDP · DAVIS, Poisson noise λ = 25 denoising | PSNR: 20.58 -> 24.63 | Improved |
| Add both spatial pyramid and variation losses (ablation) | VDP · DAVIS, Poisson noise λ = 25 denoising | PSNR: 20.58 -> 31.96 | Improved |
| Add both spatial pyramid and variation losses (ablation) | VDP · Vimeo-90KT super-resolution | PSNR: 31.23 -> 35.70 | Improved |
| Add both spatial pyramid and variation losses (ablation) | VDP · UCF-101 frame interpolation | PSNR: 33.29 -> 35.41 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 on most shared results.
- BasicVSR++ — BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond (CVPR 2021) · better on 0 of 2 shared results
- FastDVDnet — FastDVDnet: Towards Real-Time Deep Video Denoising Without Flow Estimation (CVPR 2020) · better on 1 of 2 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
- Learning Continuous Spatiotemporal Implicit Neural Fields for Unsupervised Video Denoising (TPAMI 2026 · similarity 0.88)
- Robust Video Super-Resolution With Learned Temporal Dynamics (ICCV 2017 · similarity 0.87)
- Instant Video Models: Universal Adapters for Stabilizing Image-Based Networks (NeurIPS 2025 · similarity 0.86)
- Classic Video Denoising in a Machine Learning World: Robust, Fast, and Controllable (CVPR 2025 · similarity 0.86)
- Learning Spatial Adaptation and Temporal Coherence in Diffusion Models for Video Super-Resolution (CVPR 2024 · similarity 0.85)
- FreeLong: Training-Free Long Video Generation with SpectralBlend Temporal Attention (NeurIPS 2024 · 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.
