Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression

By Zhaohui Zheng, Ping Wang, Wei Liu, Jinze Li, Rongguang Ye, Dongwei Ren. AAAI Conference on Artificial Intelligence, 2019

5,129 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/63a243afcb133569a962c41e9db956c076c5c4f3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

PP-YOLO: An Effective and Efficient Implementation of Object Detector (2020, background): “IoU loss and other variations such as CIoU loss and GIoU loss[46, 34] have been proposed to address this problem.”
YOLOv10: Real-Time End-to-End Object Detection (2024, background): “Over the past years, substantial efforts have been directed towards developing efficient detectors [19, 57, 48, 34, 79, 75, 32, 31, 41].”
Towards Large-Scale Small Object Detection: Survey and Benchmarks (2022, background): “TinaFace [132] adds a DIoU [133] branch to RetinaNet [3] and ultimately acquires a simple but strong baseline for tiny face detection.”
Slim-neck by GSConv: a lightweight-design for real-time detector architectures (2022, neutral): “As research continues to evolve, more advanced IoU loss functions have been proposed, such as GIoU [34], DIoU [35], CIoU [36] and the EIoU [37].”
Rethinking Rotated Object Detection with Gaussian Wasserstein Distance Loss (2021, methodology): “It has been shown in classic horizontal detectors that the use of IoU induced loss e.g. GIoU [45], DIoU [80] can ensure the consistency of the ﬁnal detection metric and loss.”
YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information (2024, background): “In addition, some researchers have tried to develop more general objective functions, such as loss function [5, 45, 46, 50, 77, 78], label assignment [10, 12, 33, 67, 79] and auxiliary supervision [18, 20, 24, 28, 29, 51, 54, 68, 76].”
Improved YOLOv5 network for real-time multi-scale traffic sign detection (2021, methodology): “In this paper, the GIoU is replaced by complete IoU (CIoU) [41] loss.”
YOLOv12: Attention-Centric Real-Time Object Detectors (2025, neutral): “Although improvements in YOLO have focused on areas such as loss functions [8, 35, 43, 44, 48, 67, 68], label assignment [22, 23, 34, 59, 69], network architecture design has remained a critical research priority [24, 28, 32, 57, 58].”
ActionFormer: Localizing Moments of Actions with Transformers (2022, methodology): “Our loss function, again following minimalist design, only has two terms: (1) Lcls a focal loss [36] for C way binary classification; and (2) Lreg a DIoU loss [81] for distance regression.”
Localization Distillation for Object Detection (2022, methodology): “First, we calculate the DIoU [85] matrix X between all the anchor boxes B and the ground-truth boxes B.”
YOLO-based Object Detection Models: A Review and its Applications (2024, positive): “GE uses the CIoU loss function [77] to evolve the anchor boxes by default to 1000 generations.”
Inner-IoU: More Effective Intersection over Union Loss with Auxiliary Bounding Box (2023, methodology): “The GIoU loss function [2] is defined as follows, where C is the smallest box covering B and B gt : Compared with GIoU, DIoU loss [3] function proposed to add a new distance loss term on the basis of IoU, mainly by minimizing the normalized distance between the center points of the two bounding…”
MPDIoU: A Loss for Efficient and Accurate Bounding Box Regression (2023, methodology): “Therefore, we try to design a novel loss function called LMPDIoU for bounding box regression with consideration of the advantages included in LGIoU [39], LDIoU [40], LCIoU [42], LEIoU [41], but also has higher efficiency and accuracy for bounding box regression.”
VerSe: A Vertebrae Labelling and Segmentation Benchmark for Multi-detector CT Images (2020, background): “Zheng et al. (Zheng et al., 2020) proposed CIoU loss based on these requirements.”
ASF-YOLO: A Novel YOLO Model with Attentional Scale Sequence Fusion for Cell Instance Segmentation (2023, methodology): “To address these issues, GIoU [47], DIoU, and CIoU [48] have been proposed.”
A Survey of Modern Deep Learning based Object Detection Models (2021, methodology): “YOLOv4 utilizes data augmentation techniques, regularization methods, class label smoothing, CIoU-loss [72], Cross mini-Batch Normalization (CmBN) , Self-adversarial training, Cosine annealing scheduler [73] and other tricks to improve training.”