Recent advances on loss functions in deep learning for computer vision

Y Tian, D Su, S Lauria, X Liu - Neurocomputing, 2022‏ - Elsevier
The loss function, also known as cost function, is used for training a neural network or other
machine learning models. Over the past decade, researchers have designed many loss …

Wise-IoU: bounding box regression loss with dynamic focusing mechanism

Z Tong, Y Chen, Z Xu, R Yu - arxiv preprint arxiv:2301.10051, 2023‏ - arxiv.org
The loss function for bounding box regression (BBR) is essential to object detection. Its good
definition will bring significant performance improvement to the model. Most existing works …

Mpdiou: a loss for efficient and accurate bounding box regression

S Ma, Y Xu - arxiv preprint arxiv:2307.07662, 2023‏ - arxiv.org
Bounding box regression (BBR) has been widely used in object detection and instance
segmentation, which is an important step in object localization. However, most of the existing …

UAV-YOLOv8: A small-object-detection model based on improved YOLOv8 for UAV aerial photography scenarios

G Wang, Y Chen, P An, H Hong, J Hu, T Huang - Sensors, 2023‏ - mdpi.com
Unmanned aerial vehicle (UAV) object detection plays a crucial role in civil, commercial, and
military domains. However, the high proportion of small objects in UAV images and the …

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

J He, S Erfani, X Ma, J Bailey… - Advances in Neural …, 2021‏ - proceedings.neurips.cc
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 …

Plant disease recognition model based on improved YOLOv5

Z Chen, R Wu, Y Lin, C Li, S Chen, Z Yuan, S Chen… - Agronomy, 2022‏ - mdpi.com
To accurately recognize plant diseases under complex natural conditions, an improved plant
disease-recognition model based on the original YOLOv5 network model was established …

A comprehensive review of deep learning-based PCB defect detection

X Chen, Y Wu, X He, W Ming - IEEE Access, 2023‏ - ieeexplore.ieee.org
A printed circuit board (PCB) functions as a substrate essential for interconnecting and
securing electronic components. Its widespread integration is evident in modern electronic …

Boosting R-CNN: Reweighting R-CNN samples by RPN's error for underwater object detection

P Song, P Li, L Dai, T Wang, Z Chen - Neurocomputing, 2023‏ - Elsevier
Complicated underwater environments bring new challenges to object detection, such as
unbalanced light conditions, low contrast, occlusion, and mimicry of aquatic organisms …

Rachis detection and three-dimensional localization of cut off point for vision-based banana robot

F Wu, J Duan, P Ai, Z Chen, Z Yang, X Zou - Computers and Electronics in …, 2022‏ - Elsevier
For the operation and visual positioning of a banana robot, it is important to accurately
position the rachis and cut off point. However, the main factors that affect the three …

Robust few-shot aerial image object detection via unbiased proposals filtration

L Li, X Yao, X Wang, D Hong… - IEEE Transactions on …, 2023‏ - ieeexplore.ieee.org
Few-shot aerial image object detection aims to rapidly detect object instances of novel
category in aerial images by using few labeled samples. However, due to the complex …