AI Paper Atlas

Research map / Detection and segmentation / CNN features

Deeply Unsupervised Patch Re-Identification for Pre-Training Object Detectors

Jian Ding, Enze Xie, Hang Xu, Chenhan Jiang, Zhenguo Li, Ping Luo, Gui-Song Xia

TPAMI 2022 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 41 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 patch-level re-identification loss on conv5_x (beta 0,0,0,1) (ablation)DUPR · PASCAL VOC detectionAP: 57.6 -> 58.3Improved
add patch loss on conv5_x (beta 0,0,0,1) (ablation)DUPR · COCO detection (R50-FPN)box AP: 39.2 -> 39.5Improved
add patch loss on conv4_x and conv5_x (beta 0,0,1,1) (ablation)DUPR · COCO detection (R50-FPN)box AP: 39.5 -> 40.0Improved
add multi-level image contrastive loss alpha=(0.1,0.4,0.7,1) to final-layer only (ablation)MoCo v2 · PASCAL VOC detectionAP: 57.0 -> 57.6Improved
use equal weights alpha=(1,1,1,1) on all layers (ablation)MoCo v2 · COCO Mask R-CNNbox AP: 38.9 -> 38.5Hurt
add intermediate-layer loss alpha=(0.1,0.4,0.7,1) (ablation)MoCo v2 · VOC SVM classification conv3_xaccuracy: 58.8 -> 68.8Improved
add intermediate-layer loss alpha=(0.1,0.4,0.7,1) (ablation)MoCo v2 · VOC SVM classification conv5_xaccuracy: 84.1 -> 83.7Hurt
increase RoI size from 1 to 7 for patch loss (ablation)DUPR · COCO detectionbox AP: 39.1 -> 39.5Improved
increase RoI size beyond 7 to 11 for patch loss (ablation)DUPR · VOC detectionAP: 58.3 -> 58.4No clear effect
replace image-level with patch-level contrastive loss at conv5_x (ablation)MoCo v2 · VOC detectionAP75: 63.6 -> 64.6Improved
replace image-level with patch-level contrastive loss at conv5_x (ablation)MoCo v2 · COCO detectionbox AP: 38.9 -> 38.8No clear effect
use DUPR pre-training instead of MoCo v2 with few fine-tune iterations (12k)Mask R-CNN R50-FPN · COCOmAP: +2.9 mAPImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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