Back to Search
Start Over
Improved Hard Example Mining Approach for Single Shot Object Detectors
- Publication Year :
- 2022
-
Abstract
- Hard example mining methods generally improve the performance of the object detectors, which suffer from imbalanced training sets. In this work, two existing hard example mining approaches (LRM and focal loss, FL) are adapted and combined in a state-of-the-art real-time object detector, YOLOv5. The effectiveness of the proposed approach for improving the performance on hard examples is extensively evaluated. The proposed method increases mAP by 3% compared to using the original loss function and around 1-2% compared to using the hard-mining methods (LRM or FL) individually on 2021 Anti-UAV Challenge Dataset.<br />Comment: ICIP 2022. 5 pages, 2 figures, 7 tables. The codes are available at https://github.com/aybora/yolov5Loss
- Subjects :
- Computer Science - Computer Vision and Pattern Recognition
Subjects
Details
- Database :
- arXiv
- Publication Type :
- Report
- Accession number :
- edsarx.2202.13080
- Document Type :
- Working Paper