Aspect Ratio-Based Bidirectional Label Encoding for Square-Like Rotation Detection

Rotation object detection is of great importance in remote sensing imagery where the orientation is arbitrary and objects are densely distributed. However, there are several challenges that need to be overcome, such as the angular boundary problem in the regression-based methods and the square-like problem in the classification-based methods. For square-like object rotation detection, classification-based methods [e.g., circular smooth label (CSL)] suffer from inconsistencies between angular coding and evaluation mechanisms due to the variation of aspect ratio. To address the angular inconsistencies of square-like object existing in current classification methods, we design a novel angular encoding mechanism based on aspect ratio. We make optimizations and improvements in the following two aspects: 1) proposing an aspect ratio-based bidirectional coded label (AR-BCL) to replace CSL for angle coding of square-like object, which significantly improves detection accuracy for square-like object and ii) designing a cross-fusion decoupled head (CF-DH) based on angular classification to replace the existing coupled head (CH), which can help extract features that suitable for angular classification. Extensive experiments on DOTA, a large-scale public dataset for aerial images, demonstrate the effectiveness of our method for square-like object detection.