SCALoss: Side and Corner Aligned Loss for Bounding Box Regression
This work addresses a specific bottleneck in object detection for computer vision applications, offering an incremental improvement over existing IoU-based losses.
The paper tackles the gradient vanish problem in bounding box regression for object detection by proposing SCALoss, which combines Side Overlap and Corner Distance to penalize low-overlapping cases, resulting in consistent improvements on benchmarks like COCO, PASCAL VOC, and LVIS with detectors such as YOLOV3, SSD, and Faster-RCNN.
Bounding box regression is an important component in object detection. Recent work achieves promising performance by optimizing the Intersection over Union~(IoU). However, IoU-based loss has the gradient vanish problem in the case of low overlapping bounding boxes, and the model could easily ignore these simple cases. In this paper, we propose Side Overlap~(SO) loss by maximizing the side overlap of two bounding boxes, which puts more penalty for low overlapping bounding box cases. Besides, to speed up the convergence, the Corner Distance~(CD) is added into the objective function. Combining the Side Overlap and Corner Distance, we get a new regression objective function, \textit{Side and Corner Align Loss~(SCALoss)}. The SCALoss is well-correlated with IoU loss, which also benefits the evaluation metric but produces more penalty for low-overlapping cases. It can serve as a comprehensive similarity measure, leading to better localization performance and faster convergence speed. Experiments on COCO, PASCAL VOC, and LVIS benchmarks show that SCALoss can bring consistent improvement and outperform $\ell_n$ loss and IoU based loss with popular object detectors such as YOLOV3, SSD, Faster-RCNN. Code is available at: \url{https://github.com/Turoad/SCALoss}.