Exploiting Shape Cues for Weakly Supervised Semantic Segmentation

TL;DR

This paper introduces a shape-aware approach to weakly supervised semantic segmentation that leverages shape cues and an online refinement process, significantly improving accuracy and boundary alignment over existing methods.

Contribution

It proposes a novel shape exploitation method combined with an online refinement technique, enabling end-to-end training and superior segmentation performance.

Findings

Outperforms state-of-the-art single-stage methods on PASCAL VOC 2012.

Achieves new best results among multi-stage approaches in a simple pipeline.

Produces more precise and boundary-aligned segmentation masks.

Abstract

Weakly supervised semantic segmentation (WSSS) aims to produce pixel-wise class predictions with only image-level labels for training. To this end, previous methods adopt the common pipeline: they generate pseudo masks from class activation maps (CAMs) and use such masks to supervise segmentation networks. However, it is challenging to derive comprehensive pseudo masks that cover the whole extent of objects due to the local property of CAMs, i.e., they tend to focus solely on small discriminative object parts. In this paper, we associate the locality of CAMs with the texture-biased property of convolutional neural networks (CNNs). Accordingly, we propose to exploit shape information to supplement the texture-biased CNN features, thereby encouraging mask predictions to be not only comprehensive but also well-aligned with object boundaries. We further refine the predictions in an online fashion with a novel refinement method that takes into account both the class and the color affinities, in order to generate reliable pseudo masks to supervise the model. Importantly, our model is end-to-end trained within a single-stage framework and therefore efficient in terms of the training cost. Through extensive experiments on PASCAL VOC 2012, we validate the effectiveness of our method in producing precise and shape-aligned segmentation results. Specifically, our model surpasses the existing state-of-the-art single-stage approaches by large margins. What is more, it also achieves a new state-of-the-art performance over multi-stage approaches, when adopted in a simple two-stage pipeline without bells and whistles.