COMNet: Co-Occurrent Matching for Weakly Supervised Semantic Segmentation

TL;DR

COMNet introduces a co-occurrence based matching approach that improves weakly supervised semantic segmentation by enhancing object region attention, leading to state-of-the-art results on standard datasets.

Contribution

The paper presents a novel Co-Occurrent Matching Network that enhances class activation maps by inter- and intra-matching, improving object region localization in weakly supervised segmentation.

Findings

Achieves new state-of-the-art performance on Pascal VOC 2012.

Effectively boosts baseline model performance.

Enhances object region attention in CAMs.

Abstract

Image-level weakly supervised semantic segmentation is a challenging task that has been deeply studied in recent years. Most of the common solutions exploit class activation map (CAM) to locate object regions. However, such response maps generated by the classification network usually focus on discriminative object parts. In this paper, we propose a novel Co-Occurrent Matching Network (COMNet), which can promote the quality of the CAMs and enforce the network to pay attention to the entire parts of objects. Specifically, we perform inter-matching on paired images that contain common classes to enhance the corresponded areas, and construct intra-matching on a single image to propagate the semantic features across the object regions. The experiments on the Pascal VOC 2012 and MS-COCO datasets show that our network can effectively boost the performance of the baseline model and achieve new state-of-the-art performance.