Fused-IoU Loss: Efficient Learning for Accurate Bounding Box Regression
The importance of the loss function in object detection algorithms based on deep learning has grown significantly technological progress. The accuracy of object detection is significantly affected by bounding box regression, which is a crucial factor. Since the introduction of the Intersection over...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2024-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10461559/ |
_version_ | 1797243442669027328 |
---|---|
author | Yong Sun Jianzhong Wang Hongfeng Wang Sheng Zhang Yu You Zibo Yu Yiguo Peng |
author_facet | Yong Sun Jianzhong Wang Hongfeng Wang Sheng Zhang Yu You Zibo Yu Yiguo Peng |
author_sort | Yong Sun |
collection | DOAJ |
description | The importance of the loss function in object detection algorithms based on deep learning has grown significantly technological progress. The accuracy of object detection is significantly affected by bounding box regression, which is a crucial factor. Since the introduction of the Intersection over Union (IoU) loss in 2016, many improvements have been proposed based on this loss function. These studies considered various geometric factors related to bounding boxes, and constructed penalty terms to address this issue. This paper summarizes these functions and introduces a new Fused IoU (FIoU) loss function that leads to superior performance. The FIoU loss function not only solves the problem of gradient vanishing during the backpropagation process of the IoU loss function but also solves the problem of some IoU-based loss functions degenerating into IoU loss functions under certain conditions. In addition, in the simulation experiments, the FIoU loss function resulted in faster convergence speed. In our ablation experiments across different datasets and algorithms, our aim was to compare the mAP metrics under different loss functions. On the test set of the Pascal VOC dataset, employing the Faster R-CNN algorithm, FIoU demonstrated improvements of 1.1% and 1.7% over GIoU and Smooth<inline-formula> <tex-math notation="LaTeX">$\ell _{1}$ </tex-math></inline-formula>, respectively. With the YOLOX algorithm, FIoU outperformed GIoU and IoU by 1.0% and 0.8%. Utilizing the YOLOv7 algorithm, we evaluated seven loss functions, achieving optimal results with FIoU. On the validation set of the MS-COCO 2017 dataset, using YOLOv7 and YOLOv8, FIoU exhibited gains of 0.4%, 0.2%, 0.2% over EIoU, DIoU, GIoU, and 0.3%, 0.5%, 0.3% over EIoU, DIoU, GIoU, respectively. |
first_indexed | 2024-04-24T18:55:11Z |
format | Article |
id | doaj.art-656b535d53304af9a920c68966de7b40 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-04-24T18:55:11Z |
publishDate | 2024-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-656b535d53304af9a920c68966de7b402024-03-26T17:45:06ZengIEEEIEEE Access2169-35362024-01-0112373633737710.1109/ACCESS.2024.335943310461559Fused-IoU Loss: Efficient Learning for Accurate Bounding Box RegressionYong Sun0https://orcid.org/0009-0002-5081-5905Jianzhong Wang1https://orcid.org/0000-0001-5389-4288Hongfeng Wang2Sheng Zhang3https://orcid.org/0009-0005-7727-0097Yu You4https://orcid.org/0000-0002-4261-4030Zibo Yu5https://orcid.org/0009-0009-4617-4704Yiguo Peng6School of Mechatronical Engineering, Beijing Institute of Technology, Beijing, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing, ChinaThe importance of the loss function in object detection algorithms based on deep learning has grown significantly technological progress. The accuracy of object detection is significantly affected by bounding box regression, which is a crucial factor. Since the introduction of the Intersection over Union (IoU) loss in 2016, many improvements have been proposed based on this loss function. These studies considered various geometric factors related to bounding boxes, and constructed penalty terms to address this issue. This paper summarizes these functions and introduces a new Fused IoU (FIoU) loss function that leads to superior performance. The FIoU loss function not only solves the problem of gradient vanishing during the backpropagation process of the IoU loss function but also solves the problem of some IoU-based loss functions degenerating into IoU loss functions under certain conditions. In addition, in the simulation experiments, the FIoU loss function resulted in faster convergence speed. In our ablation experiments across different datasets and algorithms, our aim was to compare the mAP metrics under different loss functions. On the test set of the Pascal VOC dataset, employing the Faster R-CNN algorithm, FIoU demonstrated improvements of 1.1% and 1.7% over GIoU and Smooth<inline-formula> <tex-math notation="LaTeX">$\ell _{1}$ </tex-math></inline-formula>, respectively. With the YOLOX algorithm, FIoU outperformed GIoU and IoU by 1.0% and 0.8%. Utilizing the YOLOv7 algorithm, we evaluated seven loss functions, achieving optimal results with FIoU. On the validation set of the MS-COCO 2017 dataset, using YOLOv7 and YOLOv8, FIoU exhibited gains of 0.4%, 0.2%, 0.2% over EIoU, DIoU, GIoU, and 0.3%, 0.5%, 0.3% over EIoU, DIoU, GIoU, respectively.https://ieeexplore.ieee.org/document/10461559/Loss functionIoUobject detectionbounding box regression |
spellingShingle | Yong Sun Jianzhong Wang Hongfeng Wang Sheng Zhang Yu You Zibo Yu Yiguo Peng Fused-IoU Loss: Efficient Learning for Accurate Bounding Box Regression IEEE Access Loss function IoU object detection bounding box regression |
title | Fused-IoU Loss: Efficient Learning for Accurate Bounding Box Regression |
title_full | Fused-IoU Loss: Efficient Learning for Accurate Bounding Box Regression |
title_fullStr | Fused-IoU Loss: Efficient Learning for Accurate Bounding Box Regression |
title_full_unstemmed | Fused-IoU Loss: Efficient Learning for Accurate Bounding Box Regression |
title_short | Fused-IoU Loss: Efficient Learning for Accurate Bounding Box Regression |
title_sort | fused iou loss efficient learning for accurate bounding box regression |
topic | Loss function IoU object detection bounding box regression |
url | https://ieeexplore.ieee.org/document/10461559/ |
work_keys_str_mv | AT yongsun fusedioulossefficientlearningforaccurateboundingboxregression AT jianzhongwang fusedioulossefficientlearningforaccurateboundingboxregression AT hongfengwang fusedioulossefficientlearningforaccurateboundingboxregression AT shengzhang fusedioulossefficientlearningforaccurateboundingboxregression AT yuyou fusedioulossefficientlearningforaccurateboundingboxregression AT ziboyu fusedioulossefficientlearningforaccurateboundingboxregression AT yiguopeng fusedioulossefficientlearningforaccurateboundingboxregression |