A Joint Framework Towards Class-aware and Class-agnostic Alignment for Few-shot Segmentation
This work addresses few-shot segmentation for computer vision, offering incremental improvements over existing methods by reducing class biases.
The paper tackles the problem of few-shot segmentation by addressing class biases and background confusion through a joint framework that combines class-aware and class-agnostic alignment guidance, achieving better segmentation performances on PASCAL-5^i and COCO-20^i datasets, particularly in the 1-shot setting.
Few-shot segmentation (FSS) aims to segment objects of unseen classes given only a few annotated support images. Most existing methods simply stitch query features with independent support prototypes and segment the query image by feeding the mixed features to a decoder. Although significant improvements have been achieved, existing methods are still face class biases due to class variants and background confusion. In this paper, we propose a joint framework that combines more valuable class-aware and class-agnostic alignment guidance to facilitate the segmentation. Specifically, we design a hybrid alignment module which establishes multi-scale query-support correspondences to mine the most relevant class-aware information for each query image from the corresponding support features. In addition, we explore utilizing base-classes knowledge to generate class-agnostic prior mask which makes a distinction between real background and foreground by highlighting all object regions, especially those of unseen classes. By jointly aggregating class-aware and class-agnostic alignment guidance, better segmentation performances are obtained on query images. Extensive experiments on PASCAL-$5^i$ and COCO-$20^i$ datasets demonstrate that our proposed joint framework performs better, especially on the 1-shot setting.