Focus More on What? Guiding Multi-Task Training for End-to-End Person Search
Abstract
End-to-end person search is a research domain that executes the task of locating and identifying a target individual from a large number of scene images via a multi-task framework. However, a major challenge for the learning of end-to-end methods is the inherent conflicts between the two sub-tasks: person detection focuses on identifying generic features of persons, while person re-identification (Re-ID) strives to find unique, distinguishing features for matching the target person. Unlike previous research focusing on model architectures, this paper delves into the end-to-end person search training process. We find that the unbalanced and conflicting training issues significantly impair the learning…
Citation impact
- FWCI
- 132.25
- Percentile
- 100%
- References
- 79
Authors
4- BCBoyu CaiCorresponding
Dalian Maritime University
- HWHuibing Wang
Dalian Maritime University
- MYMingze Yao
Dalian Maritime University
- XFXianping Fu
Dalian Maritime University
Topics & keywords
- End-to-end principle
- Focus (optics)
- Computer science
- Task (project management)
- Training (meteorology)
- Artificial intelligence
- Engineering