Causal Intervention for Weakly-Supervised Semantic Segmentation

TL;DR

This paper introduces a causal inference framework to enhance weakly-supervised semantic segmentation by reducing confounding biases, leading to more accurate pseudo-masks and improved segmentation performance.

Contribution

It proposes a structural causal model and a novel method, Context Adjustment (CONTA), to remove confounding bias in image-level classification for better segmentation.

Findings

CONTA improves segmentation accuracy on PASCAL VOC 2012

CONTA boosts performance of existing WSSS methods

Achieves new state-of-the-art results on MS-COCO

Abstract

We present a causal inference framework to improve Weakly-Supervised Semantic Segmentation (WSSS). Specifically, we aim to generate better pixel-level pseudo-masks by using only image-level labels -- the most crucial step in WSSS. We attribute the cause of the ambiguous boundaries of pseudo-masks to the confounding context, e.g., the correct image-level classification of "horse" and "person" may be not only due to the recognition of each instance, but also their co-occurrence context, making the model inspection (e.g., CAM) hard to distinguish between the boundaries. Inspired by this, we propose a structural causal model to analyze the causalities among images, contexts, and class labels. Based on it, we develop a new method: Context Adjustment (CONTA), to remove the confounding bias in image-level classification and thus provide better pseudo-masks as ground-truth for the subsequent segmentation model. On PASCAL VOC 2012 and MS-COCO, we show that CONTA boosts various popular WSSS methods to new state-of-the-arts.