Reflection Invariance Learning for Few-shot Semantic Segmentation

TL;DR

This paper introduces a novel reflection invariance learning framework for few-shot semantic segmentation, leveraging multi-view support features to improve generalization and segmentation accuracy of unseen classes.

Contribution

It proposes a multi-view matching approach that mines reflection invariance, along with modules for prior mask generation and semantic prediction, achieving state-of-the-art results.

Findings

Achieves superior performance on PASCAL-5i and COCO-20i datasets.

Effectively models reflection invariance to enhance segmentation.

Outperforms existing few-shot segmentation methods.

Abstract

Few-shot semantic segmentation (FSS) aims to segment objects of unseen classes in query images with only a few annotated support images. Existing FSS algorithms typically focus on mining category representations from the single-view support to match semantic objects of the single-view query. However, the limited annotated samples render the single-view matching struggle to perceive the reflection invariance of novel objects, which results in a restricted learning space for novel categories and further induces a biased segmentation with demoted parsing performance. To address this challenge, this paper proposes a fresh few-shot segmentation framework to mine the reflection invariance in a multi-view matching manner. Specifically, original and reflection support features from different perspectives with the same semantics are learnable fused to obtain the reflection invariance prototype with a stronger category representation ability. Simultaneously, aiming at providing better prior guidance, the Reflection Invariance Prior Mask Generation (RIPMG) module is proposed to integrate prior knowledge from different perspectives. Finally, segmentation predictions from varying views are complementarily merged in the Reflection Invariance Semantic Prediction (RISP) module to yield precise segmentation predictions. Extensive experiments on both PASCAL-$5^\textit{i}$ and COCO-$20^\textit{i}$ datasets demonstrate the effectiveness of our approach and show that our method could achieve state-of-the-art performance. Code is available at \url{https://anonymous.4open.science/r/RILFS-A4D1}