A Strong Baseline for Generalized Few-Shot Semantic Segmentation

TL;DR

This paper presents a simple yet effective generalized few-shot segmentation framework based on maximizing mutual information, improving performance on standard benchmarks with a straightforward training and inference process.

Contribution

It introduces a novel MI-based model with knowledge distillation for few-shot segmentation, applicable to any trained segmentation network, and establishes a more challenging evaluation setting.

Findings

Significant performance improvements on PASCAL-5^i and COCO-20^i benchmarks.

Gains of 7-26% in 1-shot and 3-12% in 5-shot scenarios.

Proposed a more challenging evaluation setting.

Abstract

This paper introduces a generalized few-shot segmentation framework with a straightforward training process and an easy-to-optimize inference phase. In particular, we propose a simple yet effective model based on the well-known InfoMax principle, where the Mutual Information (MI) between the learned feature representations and their corresponding predictions is maximized. In addition, the terms derived from our MI-based formulation are coupled with a knowledge distillation term to retain the knowledge on base classes. With a simple training process, our inference model can be applied on top of any segmentation network trained on base classes. The proposed inference yields substantial improvements on the popular few-shot segmentation benchmarks, PASCAL-$5^i$ and COCO-$20^i$. Particularly, for novel classes, the improvement gains range from 7% to 26% (PASCAL-$5^i$) and from 3% to 12% (COCO-$20^i$) in the 1-shot and 5-shot scenarios, respectively. Furthermore, we propose a more challenging setting, where performance gaps are further exacerbated. Our code is publicly available at https://github.com/sinahmr/DIaM.