AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Decoupled and Memory-Reinforced Networks: Towards Effective Feature Learning for One-Step Person Search

Chuchu Han, Zhedong Zheng, Changxin Gao, Nong Sang, Yi Yang

AAAI 2021 · Architectures and efficiency › RNNs and normalization · approach: Differentiable optimization layers · 46 citations (OpenAlex)

Read the paper

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
Separate detection and re-ID RoI heads (ablation)Faster R-CNN with FPN on CUHK-SYSU · CUHK-SYSURe-ID mAP: 84.3 -> 87.5Improved
Use RoI features only for identification with a one-stage detector (ablation)Faster R-CNN with FPN, then RetinaNet with FPN on CUHK-SYSU · CUHK-SYSURe-ID mAP: 87.5 -> 90.0Improved
Use ground-truth boxes only, excluding selected proposals (ablation)RetinaNet with FPN on CUHK-SYSU · CUHK-SYSURe-ID mAP: 90.0 -> 90.3Improved
Add memory-reinforced feature learning (ablation)Decoupled RetinaNet on CUHK-SYSU and PRW · CUHK-SYSU and PRWmAP: 90.3 -> 91.2; 36.1 -> 44.6Improved
Add memory-reinforced feature learning (ablation)Decoupled Reppoint on CUHK-SYSU and PRW · CUHK-SYSU and PRWmAP: 92.4 -> 92.9; 39.1 -> 46.0Improved
Increase input resolution from 1333*800 to 1500*900 (ablation)DMRNet with RetinaNet · CUHK-SYSU and PRWmAP: 91.2 -> 91.6; 44.6 -> 46.1Improved
Increase input resolution from 1333*800 to 1500*900 (ablation)DMRNet with Reppoint · CUHK-SYSU and PRWmAP: 92.9 -> 93.2; 46.0 -> 46.9Improved
Set the slow-moving encoder momentum to 0.999 rather than 0 (ablation)DMRNet on CUHK-SYSU · CUHK-SYSUmAP: 90.6 -> 91.2Improved
Use a queue rather than a look-up table for feature memory (ablation)Memory-reinforced feature learning on CUHK-SYSU · CUHK-SYSUmAP: +0.3% mAPImproved
Add more unlabeled negative instances to the queue (ablation)Memory-reinforced pairwise learning with L=2048 or L=5532 · CUHK-SYSUmAPImproved
Add more unlabeled instances to OIM (ablation)OIM loss · CUHK-SYSUmAPNo clear effect
Increase labeled queue length when the unlabeled queue is empty (ablation)Memory-reinforced feature learning · CUHK-SYSUmAPImproved

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