Grouped Adaptive Loss Weighting for Person Search

TL;DR

This paper introduces GALW, a method that automatically and dynamically adjusts task weights in multi-task person search models by grouping tasks based on convergence rates, improving optimization.

Contribution

We propose a novel grouped adaptive loss weighting approach that dynamically assigns shared weights to task groups based on loss uncertainty, addressing multi-task convergence issues in person search.

Findings

GALW improves performance on CUHK-SYSU and PRW benchmarks.

Automatically adjusts task weights without manual tuning.

Enhances multi-task learning efficiency in person search.

Abstract

Person search is an integrated task of multiple sub-tasks such as foreground/background classification, bounding box regression and person re-identification. Therefore, person search is a typical multi-task learning problem, especially when solved in an end-to-end manner. Recently, some works enhance person search features by exploiting various auxiliary information, e.g. person joint keypoints, body part position, attributes, etc., which brings in more tasks and further complexifies a person search model. The inconsistent convergence rate of each task could potentially harm the model optimization. A straightforward solution is to manually assign different weights to different tasks, compensating for the diverse convergence rates. However, given the special case of person search, i.e. with a large number of tasks, it is impractical to weight the tasks manually. To this end, we propose a Grouped Adaptive Loss Weighting (GALW) method which adjusts the weight of each task automatically and dynamically. Specifically, we group tasks according to their convergence rates. Tasks within the same group share the same learnable weight, which is dynamically assigned by considering the loss uncertainty. Experimental results on two typical benchmarks, CUHK-SYSU and PRW, demonstrate the effectiveness of our method.