Harmonious Attention Network for Person Re-Identification

CVPR 2018  ·  Wei Li, Xiatian Zhu, Shaogang Gong ·

Existing person re-identification (re-id) methods either assume the availability of well-aligned person bounding box images as model input or rely on constrained attention selection mechanisms to calibrate misaligned images. They are therefore sub-optimal for re-id matching in arbitrarily aligned person images potentially with large human pose variations and unconstrained auto-detection errors. In this work, we show the advantages of jointly learning attention selection and feature representation in a Convolutional Neural Network (CNN) by maximising the complementary information of different levels of visual attention subject to re-id discriminative learning constraints. Specifically, we formulate a novel Harmonious Attention CNN (HA-CNN) model for joint learning of soft pixel attention and hard regional attention along with simultaneous optimisation of feature representations, dedicated to optimise person re-id in uncontrolled (misaligned) images. Extensive comparative evaluations validate the superiority of this new HA-CNN model for person re-id over a wide variety of state-of-the-art methods on three large-scale benchmarks including CUHK03, Market-1501, and DukeMTMC-ReID.

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Person Re-Identification CUHK03 HA-CNN MAP 38.6 # 14
Rank-1 41.7 # 15
Person Re-Identification CUHK03 detected HA-CNN (CVPR'18) MAP 38.6 # 13
Rank-1 41.7 # 13
Person Re-Identification CUHK03 labeled HA-CNN (CVPR'18) MAP 41.0 # 16
Rank-1 44.4 # 15

Methods


No methods listed for this paper. Add relevant methods here