Single Domain Generalization for Few-Shot Counting via Universal Representation Matching

TL;DR

This paper introduces URM, a novel few-shot counting model that leverages universal vision-language representations to enhance domain generalization, achieving state-of-the-art results in unseen scenarios.

Contribution

The paper presents the first single domain generalization approach for few-shot counting, utilizing universal representations to improve robustness across diverse domains.

Findings

URM outperforms existing methods on in-domain data.

URM demonstrates superior generalization to unseen domains.

Universal representations significantly boost domain robustness.

Abstract

Few-shot counting estimates the number of target objects in an image using only a few annotated exemplars. However, domain shift severely hinders existing methods to generalize to unseen scenarios. This falls into the realm of single domain generalization that remains unexplored in few-shot counting. To solve this problem, we begin by analyzing the main limitations of current methods, which typically follow a standard pipeline that extract the object prototypes from exemplars and then match them with image feature to construct the correlation map. We argue that existing methods overlook the significance of learning highly generalized prototypes. Building on this insight, we propose the first single domain generalization few-shot counting model, Universal Representation Matching, termed URM. Our primary contribution is the discovery that incorporating universal vision-language representations distilled from a large scale pretrained vision-language model into the correlation construction process substantially improves robustness to domain shifts without compromising in domain performance. As a result, URM achieves state-of-the-art performance on both in domain and the newly introduced domain generalization setting.