Pedestrian Alignment Network for Large-scale Person Re-identification

3 Jul 2017  ·  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. read more

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Person Re-Identification CUHK03 (detected) PAN+re-rank Rank-1 41.9 # 1
MAP 43.8 # 1
Person Re-Identification CUHK03 (detected) PAN(Zheng et al., [2017a]) Rank-1 36.3 # 2
MAP 34 # 2
Person Re-Identification CUHK03 labeled PAN(Zheng et al., [2017a]) MAP 35.0 # 15
Rank-1 36.9 # 15
Person Re-Identification CUHK03 labeled PAN+re-rank MAP 45.8 # 13
Rank-1 43.9 # 14
Person Re-Identification DukeMTMC-reID PAN Rank-1 71.59 # 58
mAP 51.51 # 59
Person Re-Identification DukeMTMC-reID PAN + re-rank Rank-1 75.94 # 54
mAP 66.74 # 47
Person Re-Identification Market-1501 PAN (GAN)+re-rank Rank-1 88.57 # 59
mAP 81.53 # 53


No methods listed for this paper. Add relevant methods here