Few-Shot Segmentation Without Meta-Learning: A Good Transductive Inference Is All You Need?

TL;DR

This paper demonstrates that simple transductive inference with a novel loss function can outperform meta-learning approaches in few-shot segmentation, especially with more shots and domain shifts.

Contribution

It introduces a transductive inference method using a new loss function that improves few-shot segmentation without meta-learning, applicable with standard training.

Findings

Achieves competitive results with simple linear classifier inference.

Outperforms state-of-the-art in 5- and 10-shot scenarios on PASCAL-5i.

Performs best in domain shift settings between datasets.

Abstract

We show that the way inference is performed in few-shot segmentation tasks has a substantial effect on performances -- an aspect often overlooked in the literature in favor of the meta-learning paradigm. We introduce a transductive inference for a given query image, leveraging the statistics of its unlabeled pixels, by optimizing a new loss containing three complementary terms: i) the cross-entropy on the labeled support pixels; ii) the Shannon entropy of the posteriors on the unlabeled query-image pixels; and iii) a global KL-divergence regularizer based on the proportion of the predicted foreground. As our inference uses a simple linear classifier of the extracted features, its computational load is comparable to inductive inference and can be used on top of any base training. Foregoing episodic training and using only standard cross-entropy training on the base classes, our inference yields competitive performances on standard benchmarks in the 1-shot scenarios. As the number of available shots increases, the gap in performances widens: on PASCAL-5i, our method brings about 5% and 6% improvements over the state-of-the-art, in the 5- and 10-shot scenarios, respectively. Furthermore, we introduce a new setting that includes domain shifts, where the base and novel classes are drawn from different datasets. Our method achieves the best performances in this more realistic setting. Our code is freely available online: https://github.com/mboudiaf/RePRI-for-Few-Shot-Segmentation.