A Scale Balanced Loss for Bounding Box Regression
Object detectors typically use bounding box regressors to improve the accuracy of object localization. Currently, the two types of bounding box regression loss are ℓ<sub>n</sub>-norm-based and intersection over union (IoU)-based. However, we found that these two types of losse...
Main Authors: | Degang Sun, Yang Yang, Min Li, Jian Yang, Bo Meng, Ruwen Bai, Linghan Li, Junxing Ren |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9112187/ |
Similar Items
-
Fused-IoU Loss: Efficient Learning for Accurate Bounding Box Regression
by: Yong Sun, et al.
Published: (2024-01-01) -
Corner-Point and Foreground-Area IoU Loss: Better Localization of Small Objects in Bounding Box Regression
by: Delong Cai, et al.
Published: (2023-05-01) -
ICIoU: Improved Loss Based on Complete Intersection Over Union for Bounding Box Regression
by: Xufei Wang, et al.
Published: (2021-01-01) -
NGIoU Loss: Generalized Intersection over Union Loss Based on a New Bounding Box Regression
by: Chenghao Tong, et al.
Published: (2022-12-01) -
Bounding Box Projection for Regression Uncertainty in Oriented Object Detection
by: Qian Wu, et al.
Published: (2021-01-01)