Discovering Class-Specific Pixels for Weakly-Supervised Semantic Segmentation

TL;DR

This paper introduces a method combining saliency and attention maps to discover class-specific pixels, significantly improving weakly-supervised semantic segmentation performance on PASCAL VOC12.

Contribution

It presents a hierarchical approach to identify class-specific pixels using saliency and attention, serving as approximate ground truth for CNN training in weakly-supervised segmentation.

Findings

Achieved 60.8% mIoU on PASCAL VOC12 val set

Achieved 61.9% mIoU on PASCAL VOC12 test set

Surpassed previous state-of-the-art by over 5% mIoU

Abstract

We propose an approach to discover class-specific pixels for the weakly-supervised semantic segmentation task. We show that properly combining saliency and attention maps allows us to obtain reliable cues capable of significantly boosting the performance. First, we propose a simple yet powerful hierarchical approach to discover the class-agnostic salient regions, obtained using a salient object detector, which otherwise would be ignored. Second, we use fully convolutional attention maps to reliably localize the class-specific regions in a given image. We combine these two cues to discover class-specific pixels which are then used as an approximate ground truth for training a CNN. While solving the weakly supervised semantic segmentation task, we ensure that the image-level classification task is also solved in order to enforce the CNN to assign at least one pixel to each object present in the image. Experimentally, on the PASCAL VOC12 val and test sets, we obtain the mIoU of 60.8% and 61.9%, achieving the performance gains of 5.1% and 5.2% compared to the published state-of-the-art results. The code is made publicly available.