AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Tracing Copied Pixels and Regularizing Patch Affinity in Copy Detection

Yichen Lu, Siwei Nie, Minlong Lu, Xudong Yang, Xiaobo Zhang, Peng Zhang

ICCV 2025 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Empirical contrastive visual learning · 0 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
add CopyNCE loss (ablation)descriptor (ViT-S) · DISC21 dev IImuAP: 68.9 -> 70.5Improved
add CopyNCE loss without global hard negative mining (ablation)descriptor (ViT-S) · DISC21 dev IImuAP: 57.7 -> 61.8Improved
add CopyNCE loss to ResNet-50 descriptor (ablation)descriptor (R50, 448x448) · DISC21 dev IImuAP: 67.8 -> 68.8Improved
add CopyNCE loss on top of BCE (ablation)matcher (ViT-S) · DISC21 dev IImuAP: 70.9 -> 83.5Improved
use NCE instead of cosine loss for patch affinity (ablation)descriptor · DISC21 dev IImuAP: 68.6 -> 70.5Improved
use proportional-overlap prior (gamma=3) instead of uniform (gamma=0) (ablation)descriptor · DISC21 dev IImuAP: 67.9 -> 70.5Improved
use proportional-overlap prior (gamma=1) instead of uniform (gamma=0) (ablation)matcher · DISC21 dev IImuAP: 82.5 -> 83.5Improved
use PixTrace instead of FeatNN+LocNN matching with NCE (ablation)matcher · DISC21 dev IImuAP: 80.8 -> 83.5Improved
use PixTrace instead of FeatNN+LocNN matching with cosine (ablation)descriptor · DISC21 dev IImuAP: 68.5 -> 70.5Improved
use 4 fusion layers instead of 2 (enc-8-fus-4 vs enc-10-fus-2) (ablation)matcher · DISC21 dev IImuAP: 79.4 -> 83.5Improved
add global hard negative mining (ablation)descriptor baseline · DISC21 dev IImuAP: 57.7 -> 68.9Improved
add local crops ensemblingCopyNCE matcher ViT-S 224 · DISC21 dev IImuAP: 83.5 -> 87.4Improved

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