Reference-based Category Discovery: Unsupervised Object Detection with Category Awareness

TL;DR

RefCD is an unsupervised object detection framework that achieves category-aware detection by leveraging feature similarity with reference images, eliminating the need for manual annotations.

Contribution

It introduces a novel feature similarity loss guiding unsupervised learning of category-specific features and supports both category-aware and category-agnostic detection.

Findings

RefCD effectively learns category information without labels.

It outperforms previous unsupervised methods in category-aware detection.

RefCD can operate without reference images for category-agnostic detection.

Abstract

Traditional one-shot detection methods have addressed the closed-set problem in object detection, but the high cost of data annotation remains a critical challenge. General unsupervised methods generate pseudo boxes without category labels, thus failing to achieve category-aware classification. To overcome these limitations, we propose Reference-based Category Discovery (RefCD), an unsupervised detector that enables category-aware\footnotemark[1] detection without any manually annotated labels. It leverages feature similarity between predicted objects and unlabeled reference images. Unlike previous unsupervised methods that lack category guidance and one-shot methods which require labeled data, RefCD introduces a carefully designed feature similarity loss to explicitly guide the learning of potential category-specific features. Additionally, RefCD supports category-agnostic detection without reference images, serving as a unified framework. Comprehensive quantitative and qualitative analysis of category-aware and category-agnostic detection results demonstrates its effectiveness, and RefCD can learn category information in an unsupervised paradigm even without category labels.