AI Paper Atlas

Research map / Video / Video segmentation and tracking

Patch-VQ: 'Patching Up' the Video Quality Problem

CVPR 2021 · Video › Video segmentation and tracking · approach: Video restoration with temporal priors · 197 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 v-patch features to PVQ (ablation)PVQ · LSVQ TestSRCC: 0.814 -> 0.827Helped
Add v-patch features to PVQ (ablation)PVQ · LSVQ Test-1080pSRCC: 0.686 -> 0.711Helped
Use 3D features instead of 2D features alone (ablation)PVQ · LSVQ TestSRCC: 0.774 -> 0.805Helped
Use stv-patches rather than sv-patches alone (ablation)PVQ · LSVQ TestSRCC: 0.815 -> 0.824Helped
Use stv-patches rather than tv-patches alone (ablation)PVQ · LSVQ TestSRCC: 0.817 -> 0.824Helped
Replace the standard 2D/3D backbones with MobileNetV2 and reduce RoIPool output to 1 × 1 (ablation)PVQ · LSVQ TestSRCC: 0.827 -> 0.774Hurt
Add v-patch training (ablation)PVQ trained on LSVQ · LIVE-VQCSRCC: 0.747 -> 0.770Helped
Add v-patch training (ablation)PVQ trained on LSVQ · KoNViD-1kSRCC: 0.782 -> 0.791Helped

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