AI Paper Atlas

Research map / Video / Video segmentation and tracking

Neural Compression-Based Feature Learning for Video Restoration

Cong Huang, Jiahao Li, Bin Li, Dong Liu, Yan Lu

CVPR 2022 · Video › Video segmentation and tracking · approach: Video restoration with temporal priors · 35 citations (OpenAlex)

Check 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
Add neural compression-based feature learning to the baseline with MV refinement (ablation)Video restoration model · Set8 σ=50PSNR: 29.87 -> 30.29Improved
Add MV refinement (ablation)Video restoration baseline · Set8 σ=50PSNR: 29.75 -> 29.87Improved
Add feature attention (ablation)NCFL-based video restoration model · Set8 σ=50PSNR: 30.29 -> 30.45Improved
Use adaptive quantization instead of no quantization (ablation)NCFL · Set8 σ=50PSNR: 29.98 -> 30.29Improved
Replace adaptive quantization with a fixed quantization step (ablation)NCFL · Set8 σ=50PSNR: 30.29 -> 29.89Hurt
Add NCFL (ablation)Video deraining model · DerainingPSNR: 27.30 -> 28.11Improved
Add cross-entropy loss to NCFL (ablation)Video restoration model · DerainingPSNR: 27.64 -> 28.11Improved
Replace MV refinement with normal convolutional layers (ablation)Video restoration model · Set8 σ=50PSNR: 29.87 -> 29.79Hurt
Replace NCFL with normal convolutional layers (ablation)Video restoration model · Set8 σ=50PSNR: 30.29 -> 29.93Hurt
Increase cross-entropy loss weight to 1/256 or reduce it to 1/4096 (ablation)NCFL training · Set8 σ=50PSNR: 30.45; 30.25; 30.40Mixed
Use adaptive quantization rather than no quantization on clean features (ablation)NCFL · Clean DAVIS featuresPSNR: 30.39 -> 30.37No clear effect
Apply the full method instead of BasicVSR++Video denoising · Set8 σ=50PSNR: 30.32 -> 30.45Improved

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.

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