Adaptive Spatial-BCE Loss for Weakly Supervised Semantic Segmentation

For Weakly-Supervised Semantic Segmentation (WSSS) with image-level annotation, mostly relies on the classification network to generate initial segmentation pseudo-labels. However, the optimization target of classification networks usually neglects the discrimination between different pixels, like insignificant foreground and background regions. In this paper, we propose an adaptive Spatial Binary Cross-Entropy (Spatial-BCE) Loss for WSSS, which aims to enhance the discrimination between pixels. In Spatial-BCE Loss, we calculate the loss independently for each pixel, and heuristically assign the optimization directions for foreground and background pixels separately. An auxiliary self-supervised task is also proposed to guarantee the Spatial-BCE Loss working as envisaged. Meanwhile, to enhance the network's generalization for different data distributions, we design an alternate training strategy to adaptively generate thresholds to divide the foreground and background. Benefiting from high-quality initial pseudolabels by Spatial-BCE Loss, our method also reduce the reliance on post-processing, thereby simplifying the pipeline of WSSS. Our method is validated on the PASCAL VOC 2012 and COCO 2014 datasets, and achieves the new state-of-the-arts.

Automated summary

This paper proposes an adaptive Spatial Binary Cross-Entropy (Spatial-BCE) Loss for weakly-supervised semantic segmentation, aiming to enhance the discrimination between pixels. By calculating the loss independently for each pixel and assigning optimization directions for foreground and background pixels separately, as well as designing an alternate training strategy to generate thresholds for foreground and background, high-quality initial pseudolabels are generated, reducing the reliance on post-processing.