Exploring Visual Context for Weakly Supervised Person Search
This work addresses the scalability and practicability limitations of person search by reducing annotation costs, though it is incremental as it builds on existing weakly supervised methods.
The paper tackles weakly supervised person search, which jointly addresses pedestrian detection and re-identification using only bounding box annotations instead of full identity labels, achieving 80.0% mAP on CUHK-SYSU with an 8.8% improvement over the baseline and comparable performance to some supervised models.
Person search has recently emerged as a challenging task that jointly addresses pedestrian detection and person re-identification. Existing approaches follow a fully supervised setting where both bounding box and identity annotations are available. However, annotating identities is labor-intensive, limiting the practicability and scalability of current frameworks. This paper inventively considers weakly supervised person search with only bounding box annotations. We proposed to address this novel task by investigating three levels of context clues (i.e., detection, memory and scene) in unconstrained natural images. The first two are employed to promote local and global discriminative capabilities, while the latter enhances clustering accuracy. Despite its simple design, our CGPS achieves 80.0% in mAP on CUHK-SYSU, boosting the baseline model by 8.8%. Surprisingly, it even achieves comparable performance with several supervised person search models. Our code is available at https://github.com/ljpadam/CGPS