Learning Modulated Loss for Rotated Object Detection

TL;DR

This paper introduces a modulated loss function for rotated object detection that addresses training instability caused by angle periodicity and parameter inconsistency, achieving state-of-the-art results on multiple benchmarks.

Contribution

The paper proposes a novel rotation loss that dismisses loss discontinuity and improves regression stability, enhancing rotated object detection performance.

Findings

Achieves state-of-the-art results on DOTA and UCAS-AOD benchmarks.

Demonstrates improved generalization on ICDAR2015, HRSC2016, and FDDB.

Qualitative improvements in detection accuracy.

Abstract

Popular rotated detection methods usually use five parameters (coordinates of the central point, width, height, and rotation angle) to describe the rotated bounding box and l1-loss as the loss function. In this paper, we argue that the aforementioned integration can cause training instability and performance degeneration, due to the loss discontinuity resulted from the inherent periodicity of angles and the associated sudden exchange of width and height. This problem is further pronounced given the regression inconsistency among five parameters with different measurement units. We refer to the above issues as rotation sensitivity error (RSE) and propose a modulated rotation loss to dismiss the loss discontinuity. Our new loss is combined with the eight-parameter regression to further solve the problem of inconsistent parameter regression. Experiments show the state-of-art performances of our method on the public aerial image benchmark DOTA and UCAS-AOD. Its generalization abilities are also verified on ICDAR2015, HRSC2016, and FDDB. Qualitative improvements can be seen in Fig 1, and the source code will be released with the publication of the paper.