SFM-Net

Welcome

This repository contains official code for our paper entilted "Scale-invariant Feature Matching Network for V-D-T Few-Shot Semantic Segmentation"

Thank you for your attention.

Note: Since this paper is still in peer review, the final code will be released after it is revised according to the reviewers' comments.

Dataset

We used VDT-2048-5$^i$ to evaluate our method's performance.

Experimental Results

Before the release of our code, we will have our visualization results in the repository below.

Visual comparison with the state-of-the-art multi-modal FSS models on VDT-2048-5i dataset.

Visual validation of the reverse attention (RA) unit.

Visualization of attention weight in our k-shot strategy for a selected query patch (indicated by a red dot).