AI Paper Atlas

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

NeurIPS 2023 · Video › Video segmentation and tracking · approach: Video restoration with temporal priors · 1 citations (OpenAlex)

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 changedWhereMeasuredEffect
Optimize VDP on the test sequence rather than use a pretrained denoiserVideo denoising · DAVIS, Poisson noise λ = 25PSNR: 22.08 -> 31.96Improved
Optimize VDP on the test sequence rather than use a pretrained denoiserVideo denoising · DAVIS, Poisson noise λ = 30PSNR: 20.98 -> 30.07Improved
Use VDP instead of RIFE for interpolationVideo frame interpolation · UCF101, clean inputPSNR: 35.41 -> 35.41No clear effect
Use VDP instead of DAIN for interpolationVideo frame interpolation · UCF101, second frame noisy, σ = 15PSNR: 20.87 -> 34.92Improved
Use VDP instead of BasicVSR++ for super-resolutionVideo super-resolution · Vimeo-90K-T, clean inputPSNR: 35.95 -> 35.70Hurt
Use VDP instead of EDVR for super-resolutionVideo super-resolution · Vimeo-90K-T, noisy input σ = 5PSNR: 32.02 -> 33.87Improved
Use VDP instead of FGVC for object removalVideo object removal · DAVIS, stationary maskPSNR: 31.92 -> 32.06Improved
Add spatial pyramid loss to reconstruction loss (ablation)VDP · DAVIS, Poisson noise λ = 25 denoisingPSNR: 20.58 -> 30.14Improved
Add variation loss to reconstruction loss (ablation)VDP · DAVIS, Poisson noise λ = 25 denoisingPSNR: 20.58 -> 24.63Improved
Add both spatial pyramid and variation losses (ablation)VDP · DAVIS, Poisson noise λ = 25 denoisingPSNR: 20.58 -> 31.96Improved
Add both spatial pyramid and variation losses (ablation)VDP · Vimeo-90KT super-resolutionPSNR: 31.23 -> 35.70Improved
Add both spatial pyramid and variation losses (ablation)VDP · UCF-101 frame interpolationPSNR: 33.29 -> 35.41Improved

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy