Cascade Transformers for End-to-End Person Search

Rui Yu, Dawei Du, Rodney LaLonde, Daniel Davila, Christopher Funk, Anthony Hoogs, Brian Clipp; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 7267-7276


The goal of person search is to localize a target person from a gallery set of scene images, which is extremely challenging due to large scale variations, pose/viewpoint changes, and occlusions. In this paper, we propose the Cascade Occluded Attention Transformer (COAT) for end-to-end person search. Our three-stage cascade design focuses on detecting people in the first stage, while later stages simultaneously and progressively refine the representation for person detection and re-identification. At each stage the occluded attention transformer applies tighter intersection over union thresholds, forcing the network to learn coarse-to-fine pose/scale invariant features. Meanwhile, we calculate each detection's occluded attention to differentiate a person's tokens from other people or the background. In this way, we simulate the effect of other objects occluding a person of interest at the token-level. Through comprehensive experiments, we demonstrate the benefits of our method by achieving state-of-the-art performance on two benchmark datasets.

Related Material

[pdf] [arXiv]
@InProceedings{Yu_2022_CVPR, author = {Yu, Rui and Du, Dawei and LaLonde, Rodney and Davila, Daniel and Funk, Christopher and Hoogs, Anthony and Clipp, Brian}, title = {Cascade Transformers for End-to-End Person Search}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2022}, pages = {7267-7276} }