Dynamic Coarse-to-Fine Learning for Oriented Tiny Object Detection

TL;DR

This paper introduces a dynamic coarse-to-fine learning approach for detecting arbitrarily oriented tiny objects, addressing label assignment issues and improving performance on multiple datasets.

Contribution

It proposes a novel dynamic prior and coarse-to-fine assigner (DCFL) to improve label assignment and feature supervision for tiny oriented object detection.

Findings

Achieves state-of-the-art results on DOTA-v1.5, DOTA-v2.0, and DIOR-R datasets.

Substantial performance improvements over baseline detectors.

Effective in single-scale training and testing scenarios.

Abstract

Detecting arbitrarily oriented tiny objects poses intense challenges to existing detectors, especially for label assignment. Despite the exploration of adaptive label assignment in recent oriented object detectors, the extreme geometry shape and limited feature of oriented tiny objects still induce severe mismatch and imbalance issues. Specifically, the position prior, positive sample feature, and instance are mismatched, and the learning of extreme-shaped objects is biased and unbalanced due to little proper feature supervision. To tackle these issues, we propose a dynamic prior along with the coarse-to-fine assigner, dubbed DCFL. For one thing, we model the prior, label assignment, and object representation all in a dynamic manner to alleviate the mismatch issue. For another, we leverage the coarse prior matching and finer posterior constraint to dynamically assign labels, providing appropriate and relatively balanced supervision for diverse instances. Extensive experiments on six datasets show substantial improvements to the baseline. Notably, we obtain the state-of-the-art performance for one-stage detectors on the DOTA-v1.5, DOTA-v2.0, and DIOR-R datasets under single-scale training and testing. Codes are available at https://github.com/Chasel-Tsui/mmrotate-dcfl.