Dynamic Cascade Query Selection for Oriented Object Detection

Most of the existing object detection methods have complicated hand-designed components, such as nonmaximum suppression procedures and manual resizing of anchor boxes. Based on detection transformer (DETR), this letter not only eliminates the need for manual component adjustment but also solves three problems of poor remote sensing image for directional object capture, slow DETR convergence, and the same attention allocated by different layers of decoder. First, the D-angle module is used to align the rotating object region while accelerating the convergence using the a priori angle. Then, the overall computation of the model is reduced by using adaptive proposal selection (APS) in the cascade structure. Finally, the adaptive query selection (AQS) module is applied so that the decoder in different layers gets different attention weights to optimize the layer-by-layer fine-tuning process. In this letter, the effectiveness of the proposed method is verified using two public datasets, DOTA and HRSC2016.

Automated summary

This study proposes a method to address the issue of complicated hand-designed components in object detection using the detection transformer (DETR) framework. By utilizing the D-angle module, adaptive proposal selection, and adaptive query selection, the proposed method effectively solves the problems of capturing directional objects in remote sensing images, slow convergence of DETR, and attention allocation in the decoder.