RepMet: Representative-based metric learning for classification and one-shot object detection

TL;DR

This paper introduces RepMet, a novel end-to-end distance metric learning method that models multi-modal class distributions, improving classification and one-shot object detection performance, especially in data-scarce scenarios.

Contribution

RepMet jointly learns the backbone, embedding space, and class distributions in a single framework, advancing DML for classification and few-shot detection.

Findings

Outperforms state-of-the-art DML methods on fine-grained datasets

Achieves top results on ImageNet-LOC with few training examples

Provides a new episodic benchmark for few-shot object detection

Abstract

Distance metric learning (DML) has been successfully applied to object classification, both in the standard regime of rich training data and in the few-shot scenario, where each category is represented by only a few examples. In this work, we propose a new method for DML that simultaneously learns the backbone network parameters, the embedding space, and the multi-modal distribution of each of the training categories in that space, in a single end-to-end training process. Our approach outperforms state-of-the-art methods for DML-based object classification on a variety of standard fine-grained datasets. Furthermore, we demonstrate the effectiveness of our approach on the problem of few-shot object detection, by incorporating the proposed DML architecture as a classification head into a standard object detection model. We achieve the best results on the ImageNet-LOC dataset compared to strong baselines, when only a few training examples are available. We also offer the community a new episodic benchmark based on the ImageNet dataset for the few-shot object detection task.