Two-Phase Learning for Weakly Supervised Object Localization

TL;DR

This paper introduces a two-phase learning approach for weakly supervised object localization that progressively captures the entire object by suppressing salient parts in the second phase.

Contribution

The paper proposes a novel two-phase training scheme that enhances object localization by iteratively focusing on less salient regions, improving upon existing weakly supervised methods.

Findings

Effective in capturing entire objects rather than just salient parts

Improves performance in weakly supervised segmentation and localization tasks

Demonstrates significant gains in experimental evaluations

Abstract

Weakly supervised semantic segmentation and localiza- tion have a problem of focusing only on the most important parts of an image since they use only image-level annota- tions. In this paper, we solve this problem fundamentally via two-phase learning. Our networks are trained in two steps. In the first step, a conventional fully convolutional network (FCN) is trained to find the most discriminative parts of an image. In the second step, the activations on the most salient parts are suppressed by inference conditional feedback, and then the second learning is performed to find the area of the next most important parts. By combining the activations of both phases, the entire portion of the tar- get object can be captured. Our proposed training scheme is novel and can be utilized in well-designed techniques for weakly supervised semantic segmentation, salient region detection, and object location prediction. Detailed experi- ments demonstrate the effectiveness of our two-phase learn- ing in each task.