Singular Value Fine-tuning: Few-shot Segmentation requires Few-parameters Fine-tuning

TL;DR

This paper introduces Singular Value Fine-tuning (SVF), a novel approach for few-shot segmentation that fine-tunes only the singular values of the backbone's parameters, leading to improved generalization and state-of-the-art results.

Contribution

It proposes a new fine-tuning paradigm that adjusts only the singular values of the backbone's parameters using SVD, enhancing few-shot segmentation performance.

Findings

Achieves state-of-the-art results on Pascal-5i and COCO-20i datasets.

Outperforms existing methods in 1-shot and 5-shot settings.

Demonstrates that fine-tuning singular values effectively balances adaptation and overfitting.

Abstract

Freezing the pre-trained backbone has become a standard paradigm to avoid overfitting in few-shot segmentation. In this paper, we rethink the paradigm and explore a new regime: {\em fine-tuning a small part of parameters in the backbone}. We present a solution to overcome the overfitting problem, leading to better model generalization on learning novel classes. Our method decomposes backbone parameters into three successive matrices via the Singular Value Decomposition (SVD), then {\em only fine-tunes the singular values} and keeps others frozen. The above design allows the model to adjust feature representations on novel classes while maintaining semantic clues within the pre-trained backbone. We evaluate our {\em Singular Value Fine-tuning (SVF)} approach on various few-shot segmentation methods with different backbones. We achieve state-of-the-art results on both Pascal-5$^i$ and COCO-20$^i$ across 1-shot and 5-shot settings. Hopefully, this simple baseline will encourage researchers to rethink the role of backbone fine-tuning in few-shot settings. The source code and models will be available at https://github.com/syp2ysy/SVF.