Weakly Supervised Object Localization as Domain Adaption

TL;DR

This paper redefines weakly supervised object localization as a domain adaptation problem, introducing a novel pipeline that improves localization accuracy by aligning feature distributions across domains and utilizing target sampling strategies.

Contribution

It presents a new perspective modeling WSOL as a domain adaptation task and proposes a DA-WSOL pipeline with a target sampling strategy and DAL loss to enhance localization performance.

Findings

Outperforms state-of-the-art methods on multiple benchmarks.

Utilizes domain adaptation techniques to improve object localization.

Introduces a target sampling strategy and DAL loss for better domain alignment.

Abstract

Weakly supervised object localization (WSOL) focuses on localizing objects only with the supervision of image-level classification masks. Most previous WSOL methods follow the classification activation map (CAM) that localizes objects based on the classification structure with the multi-instance learning (MIL) mechanism. However, the MIL mechanism makes CAM only activate discriminative object parts rather than the whole object, weakening its performance for localizing objects. To avoid this problem, this work provides a novel perspective that models WSOL as a domain adaption (DA) task, where the score estimator trained on the source/image domain is tested on the target/pixel domain to locate objects. Under this perspective, a DA-WSOL pipeline is designed to better engage DA approaches into WSOL to enhance localization performance. It utilizes a proposed target sampling strategy to select different types of target samples. Based on these types of target samples, domain adaption localization (DAL) loss is elaborated. It aligns the feature distribution between the two domains by DA and makes the estimator perceive target domain cues by Universum regularization. Experiments show that our pipeline outperforms SOTA methods on multi benchmarks. Code are released at \url{https://github.com/zh460045050/DA-WSOL_CVPR2022}.