FreeAnchor: Learning to Match Anchors for Visual Object Detection

TL;DR

FreeAnchor introduces a flexible, learning-based anchor matching method for CNN object detectors, breaking traditional IoU restrictions and improving detection accuracy through maximum likelihood estimation.

Contribution

It proposes a novel learning-to-match approach that replaces fixed anchor assignment with a trainable, likelihood-based method, enhancing detection performance.

Findings

Outperforms existing methods on COCO dataset

Significant accuracy improvements over traditional anchor assignment

Compatible with various CNN-based detectors

Abstract

Modern CNN-based object detectors assign anchors for ground-truth objects under the restriction of object-anchor Intersection-over-Unit (IoU). In this study, we propose a learning-to-match approach to break IoU restriction, allowing objects to match anchors in a flexible manner. Our approach, referred to as FreeAnchor, updates hand-crafted anchor assignment to "free" anchor matching by formulating detector training as a maximum likelihood estimation (MLE) procedure. FreeAnchor targets at learning features which best explain a class of objects in terms of both classification and localization. FreeAnchor is implemented by optimizing detection customized likelihood and can be fused with CNN-based detectors in a plug-and-play manner. Experiments on COCO demonstrate that FreeAnchor consistently outperforms their counterparts with significant margins.