期刊
IEEE TRANSACTIONS ON CYBERNETICS
卷 52, 期 8, 页码 8574-8586出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCYB.2021.3095305
关键词
Object detection; Training; Graphics processing units; Real-time systems; Detectors; Testing; Performance gain; Bounding-box regression; instance segmentation; nonmaximum suppression (NMS); object detection
类别
资金
- National Natural Science Foundation of China [61801326, U19A2073]
The proposed CIoU loss and Cluster-NMS approach, which incorporates geometric factors, significantly improve average precision and average recall in object detection and instance segmentation, with notable gains without sacrificing inference efficiency.
Deep learning-based object detection and instance segmentation have achieved unprecedented progress. In this article, we propose complete-IoU (CIoU) loss and Cluster-NMS for enhancing geometric factors in both bounding-box regression and nonmaximum suppression (NMS), leading to notable gains of average precision (AP) and average recall (AR), without the sacrifice of inference efficiency. In particular, we consider three geometric factors, that is: 1) overlap area; 2) normalized central-point distance; and 3) aspect ratio, which are crucial for measuring bounding-box regression in object detection and instance segmentation. The three geometric factors are then incorporated into CIoU loss for better distinguishing difficult regression cases. The training of deep models using CIoU loss results in consistent AP and AR improvements in comparison to widely adopted l $_{n}$ -norm loss and IoU-based loss. Furthermore, we propose Cluster-NMS, where NMS during inference is done by implicitly clustering detected boxes and usually requires fewer iterations. Cluster-NMS is very efficient due to its pure GPU implementation, and geometric factors can be incorporated to improve both AP and AR. In the experiments, CIoU loss and Cluster-NMS have been applied to state-of-the-art instance segmentation (e.g., YOLACT and BlendMask-RT), and object detection (e.g., YOLO v3, SSD, and Faster R-CNN) models. Taking YOLACT on MS COCO as an example, our method achieves performance gains as +1.7 AP and +6.2 AR $_{100}$ for object detection, and +1.1 AP and +3.5 AR $_{100}$ for instance segmentation, with 27.1 FPS on one NVIDIA GTX 1080Ti GPU. All the source code and trained models are available at https://github.com/Zzh-tju/CIoU.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据