Visual Prompting for Generalized Few-shot Segmentation: A Multi-scale Approach

TL;DR

This paper introduces a multi-scale visual prompting method for generalized few-shot semantic segmentation using transformer decoders, achieving state-of-the-art results without test-time optimization by learning prompts and employing a causal attention mechanism.

Contribution

It proposes a novel visual prompting approach with a unidirectional causal attention mechanism for dense segmentation, improving performance on both novel and base categories in few-shot settings.

Findings

State-of-the-art GFSS performance on COCO-20i and Pascal-5i datasets.

Effective use of learned visual prompts without test-time optimization.

Transductive prompt tuning further enhances segmentation accuracy.

Abstract

The emergence of attention-based transformer models has led to their extensive use in various tasks, due to their superior generalization and transfer properties. Recent research has demonstrated that such models, when prompted appropriately, are excellent for few-shot inference. However, such techniques are under-explored for dense prediction tasks like semantic segmentation. In this work, we examine the effectiveness of prompting a transformer-decoder with learned visual prompts for the generalized few-shot segmentation (GFSS) task. Our goal is to achieve strong performance not only on novel categories with limited examples, but also to retain performance on base categories. We propose an approach to learn visual prompts with limited examples. These learned visual prompts are used to prompt a multiscale transformer decoder to facilitate accurate dense predictions. Additionally, we introduce a unidirectional causal attention mechanism between the novel prompts, learned with limited examples, and the base prompts, learned with abundant data. This mechanism enriches the novel prompts without deteriorating the base class performance. Overall, this form of prompting helps us achieve state-of-the-art performance for GFSS on two different benchmark datasets: COCO-$20^i$ and Pascal-$5^i$, without the need for test-time optimization (or transduction). Furthermore, test-time optimization leveraging unlabelled test data can be used to improve the prompts, which we refer to as transductive prompt tuning.