Learning High-Precision Bounding Box for Rotated Object Detection via Kullback-Leibler Divergence

TL;DR

This paper introduces a novel rotation regression loss based on Kullback-Leibler Divergence, enabling high-precision rotated object detection by adaptively modulating parameter importance, especially for objects with large aspect ratios.

Contribution

It proposes converting rotated bounding boxes into 2-D Gaussian distributions and using KLD as a loss, improving high-precision detection and addressing limitations of existing regression losses.

Findings

KLD-based loss improves detection accuracy on seven datasets.

The method dynamically adjusts gradient importance based on object aspect ratio.

KLD loss is proven to be scale invariant and degenerates to $l_{n}$-norm loss for horizontal detection.

Abstract

Existing rotated object detectors are mostly inherited from the horizontal detection paradigm, as the latter has evolved into a well-developed area. However, these detectors are difficult to perform prominently in high-precision detection due to the limitation of current regression loss design, especially for objects with large aspect ratios. Taking the perspective that horizontal detection is a special case for rotated object detection, in this paper, we are motivated to change the design of rotation regression loss from induction paradigm to deduction methodology, in terms of the relation between rotation and horizontal detection. We show that one essential challenge is how to modulate the coupled parameters in the rotation regression loss, as such the estimated parameters can influence to each other during the dynamic joint optimization, in an adaptive and synergetic way. Specifically, we first convert the rotated bounding box into a 2-D Gaussian distribution, and then calculate the Kullback-Leibler Divergence (KLD) between the Gaussian distributions as the regression loss. By analyzing the gradient of each parameter, we show that KLD (and its derivatives) can dynamically adjust the parameter gradients according to the characteristics of the object. It will adjust the importance (gradient weight) of the angle parameter according to the aspect ratio. This mechanism can be vital for high-precision detection as a slight angle error would cause a serious accuracy drop for large aspect ratios objects. More importantly, we have proved that KLD is scale invariant. We further show that the KLD loss can be degenerated into the popular $l_{n}$-norm loss for horizontal detection. Experimental results on seven datasets using different detectors show its consistent superiority, and codes are available at https://github.com/yangxue0827/RotationDetection and https://github.com/open-mmlab/mmrotate.