Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression
Open accessAbstract
Bounding box regression is the crucial step in object detection. In existing methods, while ℓn-norm loss is widely adopted for bounding box regression, it is not tailored to the evaluation metric, i.e., Intersection over Union (IoU). Recently, IoU loss and generalized IoU (GIoU) loss have been proposed to benefit the IoU metric, but still suffer from the problems of slow convergence and inaccurate regression. In this paper, we propose a Distance-IoU (DIoU) loss by incorporating the normalized distance between the predicted box and the target box, which converges much faster in training than IoU and GIoU losses. Furthermore, this paper summarizes three geometric factors in bounding box regression, i.e., overlap area, central point distance and aspect ratio, based on which a Complete IoU (CIoU) loss is proposed, thereby leading to faster convergence and better performance. By incorporating DIoU and CIoU losses into state-of-the-art object detection algorithms, e.g., YOLO v3, SSD and Faster R-CNN, we achieve notable performance gains in terms of not only IoU metric but also GIoU metric. Moreover, DIoU can be easily adopted into non-maximum suppression (NMS) to act as the criterion, further boosting performance improvement. The source code and trained models are available at https://github.com/Zzh-tju/DIoU.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Edge YOLO has a faster convergence speed due to its relatively simple network model and the introduction of DIoU for loss calculation.
“Edge YOLO owns a relatively simple network model. DIoU [49] is introduced to calculate Loss, and the speed of convergence is faster.”
From Edge YOLO: Real-Time Intelligent Object Detection System Based on Edge-Cloud Cooperation in Autonomous Vehicles (2022) · cites this paper as [49]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 56,035 · see citing works
You Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · see citing works
Girshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105 · see citing works
FCOS: Fully Convolutional One-Stage Object Detection
Tian, Z., Shen, C., Chen, H., & He, T. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00972 · Cited by 7,013 · see citing works
Adaptive NMS: Refining Pedestrian Detection in a Crowd
Liu, S., Huang, D., & Wang, Y. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00662 · Cited by 413 · see citing works
Flagged first, then by citations
Research works printed but not linked: 26
- Bae, S.-H. 2019. Object detection based on region decomposition and assembly. In The AAAI Conference on Artificial Intelligence . Bodla, N.; Singh, B.; Chellappa, R.; and Davis, L. S. 2017. Softnms - improving object detection with one line of code. In The IEEE International Conference on Computer Vision (ICCV) .
- Cai, Z., and Vasconcelos, N. 2018. Cascade r-cnn: Delving into high quality object detection. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
- Cui, Y.; Jia, M.; Lin, T.-Y .; Song, Y .; and Belongie, S. 2019. Classbalanced loss based on effective number of samples. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
- Everingham, M.; Van Gool, L.; Williams, C. K. I.; Winn, J.; and Zisserman, A. 2010. The pascal visual object classes (voc) challenge. International Journal of Computer Vision 88(2):303-338.
- Fu, C.-Y.; Liu, W.; Ranga, A.; Tyagi, A.; and Berg, A. C. 2017. DSSD: Deconvolutional single shot detector. arXiv:1701.06659 . · arXiv
- Girshick, R.; Donahue, J.; Darrell, T.; and Malik, J. 2014. Rich feature hierarchies for accurate object detection and semantic segmentation. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
- Girshick, R. 2015. Fast r-cnn. In The IEEE International Conference on Computer Vision (ICCV) .
- He, K.; Gkioxari, G.; Dollar, P.; and Girshick, R. 2017. Mask rcnn. In The IEEE International Conference on Computer Vision (ICCV) .
- He, Y.; Zhu, C.; Wang, J.; Savvides, M.; and Zhang, X. 2019. Bounding box regression with uncertainty for accurate object detection. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
- Jiang, B.; Luo, R.; Mao, J.; Xiao, T.; and Jiang, Y. 2018. Acquisition of localization confidence for accurate object detection. In The European Conference on Computer Vision (ECCV) .
- Law, H., and Deng, J. 2018. Cornernet: Detecting objects as paired keypoints. In The European Conference on Computer Vision (ECCV) .
- Li, B.; Liu, Y.; and Wang, X. 2019. Gradient harmonized singlestage detector. In The AAAI Conference on Artificial Intelligence .
- Lin, T.-Y.; Maire, M.; Belongie, S.; Hays, J.; Perona, P.; Ramanan, D.; Doll´ ar, P.; and Zitnick, C. L. 2014. Microsoft coco: Common objects in context. In The European Conference on Computer Vision (ECCV) .
- Lin, T.-Y.; Goyal, P.; Girshick, R.; He, K.; and Dollar, P. 2017. Focal loss for dense object detection. In The IEEE International Conference on Computer Vision (ICCV) .
- Liu, W.; Anguelov, D.; Erhan, D.; Szegedy, C.; Reed, S.; Fu, C.-Y.; and Berg, A. C. 2016. Ssd: Single shot multibox detector. In The European Conference on Computer Vision (ECCV) .
- Pang, J.; Chen, K.; Shi, J.; Feng, H.; Ouyang, W.; and Lin, D. 2019. Libra r-cnn: Towards balanced learning for object detection. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
- Redmon, J., and Farhadi, A. 2017. Yolo9000: Better, faster, stronger. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
- Redmon, J., and Farhadi, A. 2018. Yolov3: An incremental improvement. arXiv:1804.02767 . · arXiv
- Rezatofighi, H.; Tsoi, N.; Gwak, J.; Sadeghian, A.; Reid, I.; and Savarese, S. 2019. Generalized intersection over union: A metric and a loss for bounding box regression. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
- Song, T.; Sun, L.; Xie, D.; Sun, H.; and Pu, S. 2018. Small-scale pedestrian detection based on topological line localization and temporal feature aggregation. In The European Conference on Computer Vision (ECCV) .
- Wang, H.; Wang, Q.; Gao, M.; Li, P.; and Zuo, W. 2018. Multiscale location-aware kernel representation for object detection. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
- Wang, H.; Wang, Q.; Yang, F.; Zhang, W.; and Zuo, W. 2019. Data augmentation for object detection via progressive and selective instance-switching. arXiv:1906.00358 . · arXiv
- Yang, T.; Zhang, X.; Li, Z.; Zhang, W.; and Sun, J. 2018. Metaanchor: Learning to detect objects with customized anchors. In Advances in Neural Information Processing Systems .
- Yang, Z.; Liu, S.; Hu, H.; Wang, L.; and Lin, S. 2019. Reppoints: Point set representation for object detection. In The IEEE International Conference on Computer Vision (ICCV) .
- Yu, J.; Jiang, Y .; Wang, Z.; Cao, Z.; and Huang, T. 2016. Unitbox: An advanced object detection network. In Proceedings of the ACM International Conference on Multimedia .
- Zhu, C.; He, Y.; and Savvides, M. 2019. Feature selective anchorfree module for single-shot object detection. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) .
Cited by
Sorted by citations
YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors
Wang, C.-Y., Bochkovskiy, A., & Liao, H.-Y. M. (2023) · 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52729.2023.00721 · Cited by 13,198 · see citing works
“Being able to become a stateof-the-art real-time object detector usually requires the following characteristics: (1) a faster and stronger network architecture; (2) a more effective feature integration method [8, 21, 29, 36, 43, 56, 71, 94]; (3) a more accurate detection method [66, 73, 74]; (4) a more robust loss fun…”
MethodYOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
Wang, C.-Y., Bochkovskiy, A., & Liao, H.-Y. M. (2022) · arXiv · doi:10.48550/arxiv.2207.02696 · Cited by 13,205 · see citing works
“Being able to become a stateof-the-art real-time object detector usually requires the following characteristics: (1) a faster and stronger network architecture; (2) a more effective feature integration method [8, 21, 29, 36, 43, 56, 71, 94]; (3) a more accurate detection method [66, 73, 74]; (4) a more robust loss fun…”
MethodYOLOv4: Optimal Speed and Accuracy of Object Detection
Bochkovskiy, A., Wang, C.-Y., & Liao, H.-Y. M. (2020) · arXiv · doi:10.48550/arxiv.2004.10934 · Cited by 10,307 · see citing works
Object Detection in 20 Years: A Survey
Zou, Z., Chen, K., Shi, Z., Guo, Y., & Ye, J. (2023) · Proceedings of the IEEE · doi:10.1109/jproc.2023.3238524 · Cited by 3,689 · see citing works
YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications
Li, C., Li, L., Jiang, H., Weng, K., Geng, Y., Li, L., Ke, Z., Li, Q., Cheng, M., Nie, W., et al. (2022) · arXiv · doi:10.48550/arxiv.2209.02976 · Cited by 1,792 · see citing works
“There are many variants of IoU, such as GIoU [35], DIoU [52], CIoU [52], α-IoU [11] and SIoU [8], etc, forming relevant loss functions.”
BackgroundInfluentialZheng, Z., Wang, P., Ren, D., Liu, W., Ye, R., Hu, Q., & Zuo, W. (2022) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2021.3095305 · Cited by 1,552 · see citing works
Self-citationHussain, M. (2023) · Machines · doi:10.3390/machines11070677 · Cited by 1,217 · see citing works
Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking
Cao, J., Pang, J., Weng, X., Khirodkar, R., & Kitani, K. (2023) · 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52729.2023.00934 · Cited by 1,115 · see citing works
A survey and performance evaluation of deep learning methods for small object detection
Liu, Y., Sun, P., Wergeles, N., & Shang, Y. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.114602 · Cited by 665 · see citing works
YOLOP: You Only Look Once for Panoptic Driving Perception
Wu, D., Liao, M.-W., Zhang, W.-T., Wang, X.-G., Bai, X., Cheng, W.-Q., & Liu, W.-Y. (2022) · Machine Intelligence Research · doi:10.1007/s11633-022-1339-y · Cited by 466 · see citing works
Self-citationLiang, S., Wu, H., Zhen, L., Hua, Q., Garg, S., Kaddoum, G., Hassan, M. M., & Yu, K. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2022.3158253 · Cited by 367 · see citing works
Robust multi-task learning and online refinement for spacecraft pose estimation across domain gap
Park, T. H., & D’Amico, S. (2024) · Advances in Space Research · doi:10.1016/j.asr.2023.03.036 · Cited by 125 · see citing works
PP-PicoDet: A Better Real-Time Object Detector on Mobile Devices
Yu, G., Chang, Q., Lv, W., Xu, C., Cui, C., Ji, W., Dang, Q., Deng, K., Wang, G., Du, Y., et al. (2021) · arXiv · doi:10.48550/arxiv.2111.00902 · Cited by 112 · see citing works
“Specifically, we use the weighted sum of Varifocal Loss (VFL)[8] and GIoU loss[9] to calculate the cost matrix, enhancing accuracy without harming efficiency.”
MethodCheng, Y., Xu, H., & Liu, Y. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01498 · Cited by 109 · see citing works
A High-Precision Vehicle Detection and Tracking Method Based on the Attention Mechanism
Wang, J., Dong, Y., Zhao, S., & Zhang, Z. (2023) · Sensors · doi:10.3390/s23020724 · Cited by 60 · see citing works
Zhu, J., Li, X., Jin, P., Xu, Q., Sun, Z., & Song, X. (2020) · Sensors · doi:10.3390/s21010027 · Cited by 51 · see citing works
“The overall processes are summarized as following: lossab = ∑ n ∑ l ( θ ∗ losslf ocal + (1− θ) ∗ loss l DIOU ) (2) l∗ = argmin l ( β ∗ losslf ocal + (1− β) ∗ loss l DIOU ) (3) lossa f = ∑ n ( β ∗ lossl∗f ocal + (1− β) ∗ loss l∗ DIOU ) (4) lossen = α ∗ lossa f + (1− α) ∗ lossab (5) After integrating the anchor-based an…”
BackgroundIntelligent detection of fastener defects in ballastless tracks based on deep learning
Ye, W., Ren, J., Lu, C., Zhang, A. A., Zhan, Y., & Liu, J. (2024) · Automation in Construction · doi:10.1016/j.autcon.2024.105280 · Cited by 42 · see citing works
PVSPE: A pyramid vision multitask transformer network for spacecraft pose estimation
Yang, H., Xiao, X., Yao, M., Xiong, Y., Cui, H., & Fu, Y. (2024) · Advances in Space Research · doi:10.1016/j.asr.2024.05.011 · Cited by 41 · see citing works
Qiao, S., Pang, S., Luo, G., Pan, S., Chen, T., & Lv, Z. (2022) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2021.3091579 · Cited by 41 · see citing works
Adverse Weather Target Detection Algorithm Based on Adaptive Color Levels and Improved YOLOv5
Yao, J., Fan, X., Li, B., & Qin, W. (2022) · Sensors · doi:10.3390/s22218577 · Cited by 35 · see citing works
Improved YOLOv8-Based Target Precision Detection Algorithm for Train Wheel Tread Defects
Wen, Y., Gao, X., Luo, L., & Li, J. (2024) · Sensors · doi:10.3390/s24113477 · Cited by 18 · see citing works
Self-citationSpacecraft Homography Pose Estimation with Single-Stage Deep Convolutional Neural Network
Chen, S., Yang, W., Wang, W., Mai, J., Liang, J., & Zhang, X. (2024) · Sensors · doi:10.3390/s24061828 · Cited by 12 · see citing works
- Authors
- Zheng, Z., Wang, P., Liu, W., Li, J., Ye, R., & Ren, D.
- Affiliations
- Tianjin University (CN)
- China People's Public Security University (CN)
- Venue
- Proceedings of the AAAI Conference on Artificial Intelligence
- Published
- 2020-04-03
- Volume
- 34
- Issue
- 07
- Pages
- 12993-13000
- ISSN
- 2159-5399 (ISSN-L)
- 2374-3468
- OpenAlex
- W2997747012
- Cited by
- 5,288
- References
- 5
- Citations by year
- 2026: 638 · 2025: 1203 · 2024: 1340 · 2023: 1013 · 2022: 656 · 2021: 367 · 2020: 71
- FWCI
- 162.49
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- www.aaai.org/ (unspecified)
- Locations
- Proceedings of the AAAI Conference on Artificial Intelligence (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Natural Science Foundation of China (91746107, 61801326)
- People's Public Security University of China
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- bounding box regression, IoU loss, object detection, loss function design, convergence rate, aspect ratio, non-maximum suppression, YOLOv3, Faster R-CNN, SSD
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
