Alpha-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression

By Jiabo He, Sarah Monazam Erfani, Xingjun Ma, James Bailey, Ying Chi, Xiansheng Hua. 2021

333 citations recorded by the source corpus.

Bounding box (bbox) regression is a fundamental task in computer vision. So far, the most commonly used loss functions for bbox regression are the Intersection over Union (IoU) loss and its variants. In this paper, we generalize existing IoU-based losses to a new family of power IoU losses that have a power IoU term and an additional power regularization term with a single power parameter $\alpha$. We call this new family of losses the $\alpha$-IoU losses and analyze properties such as order preservingness and loss/gradient reweighting. Experiments on multiple object detection benchmarks and models demonstrate that $\alpha$-IoU losses, 1) can surpass existing IoU-based losses by a noticeable performance margin; 2) offer detectors more flexibility in achieving different levels of bbox regression accuracy by modulating $\alpha$; and 3) are more robust to small datasets and noisy bboxes.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

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

1. Liu Liu
2. Dacheng Tao
3. Zhen Wang
4. Yiqun Duan
5. Xiangxiang Chu
6. Qingyuan Li
7. Liang Li
8. Xiaolin Wei

In Computer Vision and Pattern Recognition

1. Liu Liu
2. Dacheng Tao
3. Zhen Wang
4. Xiangxiang Chu
5. Qingyuan Li
6. Xiaolin Wei
7. Chuyin Li
8. Yiqun Duan

Citations in the corpus

• Continual Learning through Retrieval and Imagination (2022, background): “Metric learning losses (Hoffer et al. 2015; He et al. 2021; Wang et al. 2019a) are used for training embedding networks, with the aim to reduce the distance between similar data and to increase the distance between dissimilar data.”
• YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications (2022, background): “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.”