Long-tail Detection with Effective Class-Margins

TL;DR

This paper introduces Effective Class-Margin Loss (ECM), a novel training objective for long-tail object detection that improves performance on rare classes by optimizing a margin-based classification error, supported by theoretical analysis.

Contribution

The paper provides a theoretical framework linking mean average precision to margin-based classification error and proposes ECM loss to enhance long-tail detection.

Findings

ECM outperforms existing methods on LVIS v1 benchmark.

Theoretical analysis connects detection metrics to classification margins.

ECM is simple, effective, and broadly applicable.

Abstract

Large-scale object detection and instance segmentation face a severe data imbalance. The finer-grained object classes become, the less frequent they appear in our datasets. However, at test-time, we expect a detector that performs well for all classes and not just the most frequent ones. In this paper, we provide a theoretical understanding of the long-trail detection problem. We show how the commonly used mean average precision evaluation metric on an unknown test set is bound by a margin-based binary classification error on a long-tailed object detection training set. We optimize margin-based binary classification error with a novel surrogate objective called \textbf{Effective Class-Margin Loss} (ECM). The ECM loss is simple, theoretically well-motivated, and outperforms other heuristic counterparts on LVIS v1 benchmark over a wide range of architecture and detectors. Code is available at \url{https://github.com/janghyuncho/ECM-Loss}.