Spatial Structure Constraints for Weakly Supervised Semantic Segmentation

Tao Chen

doi:10.1109/TIP.2024.3359041

Spatial Structure Constraints for Weakly Supervised Semantic Segmentation

2024-01-01

6

OA

AI

DOI：10.1109/TIP.2024.3359041

原文链接

原文求助

摘要

En 中文

The image-level label has prevailed in weakly supervised semantic segmentation tasks due to its easy availability. Since image-level labels can only indicate the existence or absence of specific categories of objects, visualization-based techniques have been widely adopted to provide object location clues. Considering class activation maps (CAMs) can only locate the most discriminative part of objects, recent approaches usually adopt an expansion strategy to enlarge the activation area for more integral object localization. However, without proper constraints, the expanded activation will easily intrude into the background region. In this paper, we propose spatial structure constraints (SSC) for weakly supervised semantic segmentation to alleviate the unwanted object over-activation of attention expansion. Specifically, we propose a CAM-driven reconstruction module to directly reconstruct the input image from deep CAM features, which constrains the diffusion of last-layer object attention by preserving the coarse spatial structure of the image content. Moreover, we propose an activation self-modulation module to refine CAMs with finer spatial structure details by enhancing regional consistency. Without external saliency models to provide background clues, our approach achieves 72.7% and 47.0% mIoU on the PASCAL VOC 2012 and COCO datasets, respectively, demonstrating the superiority of our proposed approach.

Keyword:

Cams

Semantic segmentation

Image reconstruction

Training

Task analysis

Annotations

Semantics

weak supervision

image-level label

spatial structure constraints

Spatial Structure Constraints for Weakly Supervised Semantic Segmentation

摘要

期刊

机构