Papers › Pedestrian Alignment Network for Large-scale Person Re-identification

Pedestrian Alignment Network for Large-scale Person Re-identification

3 Jul 2017arXiv:1707.00408archive 2025-07-28

Zhedong Zheng, Liang Zheng, Yi Yang

Person re-identification (person re-ID) is mostly viewed as an image retrieval problem. This task aims to search a query person in a large image pool. In practice, person re-ID usually adopts automatic detectors to obtain cropped pedestrian images. However, this process suffers from two types of detector errors: excessive background and part missing. Both errors deteriorate the quality of pedestrian alignment and may compromise pedestrian matching due to the position and scale variances. To address the misalignment problem, we propose that alignment can be learned from an identification procedure. We introduce the pedestrian alignment network (PAN) which allows discriminative embedding learning and pedestrian alignment without extra annotations. Our key observation is that when the convolutional neural network (CNN) learns to discriminate between different identities, the learned feature maps usually exhibit strong activations on the human body rather than the background. The proposed network thus takes advantage of this attention mechanism to adaptively locate and align pedestrians within a bounding box. Visual examples show that pedestrians are better aligned with PAN. Experiments on three large-scale re-ID datasets confirm that PAN improves the discriminative ability of the feature embeddings and yields competitive accuracy with the state-of-the-art methods.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

layumi/Pedestrian_Alignment officialmentioned in papermentioned on GitHub report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image RetrievalLarge-Scale Person Re-IdentificationPerson Re-IdentificationRetrieval

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Person Re-Identification CUHK03 (detected) PAN+re-rank MAP 43.8 #1 of 2 Archive leaderboard report
Person Re-Identification CUHK03 (detected) PAN+re-rank Rank-1 41.9 #1 of 2 Archive leaderboard report
Person Re-Identification CUHK03 (detected) PAN(Zheng et al., [2017a]) MAP 34 #2 of 2 Archive leaderboard report
Person Re-Identification CUHK03 (detected) PAN(Zheng et al., [2017a]) Rank-1 36.3 #2 of 2 Archive leaderboard report
Person Re-Identification CUHK03 labeled PAN+re-rank MAP 45.8 #15 of 21 Archive leaderboard report
Person Re-Identification CUHK03 labeled PAN+re-rank Rank-1 43.9 #15 of 21 Archive leaderboard report
Person Re-Identification CUHK03 labeled PAN(Zheng et al., [2017a]) MAP 35.0 #17 of 21 Archive leaderboard report
Person Re-Identification CUHK03 labeled PAN(Zheng et al., [2017a]) Rank-1 36.9 #17 of 21 Archive leaderboard report
Person Re-Identification DukeMTMC-reID PAN + re-rank Rank-1 75.94 #69 of 94 Archive leaderboard report
Person Re-Identification DukeMTMC-reID PAN + re-rank mAP 66.74 #69 of 94 Archive leaderboard report
Person Re-Identification DukeMTMC-reID PAN Rank-1 71.59 #81 of 94 Archive leaderboard report
Person Re-Identification DukeMTMC-reID PAN mAP 51.51 #81 of 94 Archive leaderboard report
Person Re-Identification Market-1501 PAN (GAN)+re-rank Rank-1 88.57 #98 of 135 Archive leaderboard report
Person Re-Identification Market-1501 PAN (GAN)+re-rank mAP 81.53 #98 of 135 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections