Memory-guided Network with Uncertainty-based Feature Augmentation for Few-shot Semantic Segmentation

TL;DR

This paper introduces a memory-guided network with uncertainty-based feature augmentation to enhance few-shot semantic segmentation, effectively improving generalization to novel classes with limited data.

Contribution

The paper proposes a class-shared memory module and an uncertainty-based feature augmentation method to better align distributions and improve robustness in few-shot segmentation.

Findings

Outperforms state-of-the-art on PASCAL-5i dataset

Achieves superior results on COCO-20i dataset

Enhances model robustness with diverse feature generation

Abstract

The performance of supervised semantic segmentation methods highly relies on the availability of large-scale training data. To alleviate this dependence, few-shot semantic segmentation (FSS) is introduced to leverage the model trained on base classes with sufficient data into the segmentation of novel classes with few data. FSS methods face the challenge of model generalization on novel classes due to the distribution shift between base and novel classes. To overcome this issue, we propose a class-shared memory (CSM) module consisting of a set of learnable memory vectors. These memory vectors learn elemental object patterns from base classes during training whilst re-encoding query features during both training and inference, thereby improving the distribution alignment between base and novel classes. Furthermore, to cope with the performance degradation resulting from the intra-class variance across images, we introduce an uncertainty-based feature augmentation (UFA) module to produce diverse query features during training for improving the model's robustness. We integrate CSM and UFA into representative FSS works, with experimental results on the widely-used PASCAL-5$^i$ and COCO-20$^i$ datasets demonstrating the superior performance of ours over state of the art.