What is it about?
In the field of automated driving, Given the rapid advancement of technology, the demand for 3D target datasets is also increasing, and traditional manual labeling is too costly and inefficient. This paper first introduces the methods of 3D automated labeling based on deep learning, LiDAR, and SAM. This paper provides an overview of LIDAR and SAM methods, as well as a brief introduction to the principles of representative, cutting-edge work in related fields.
Featured Image
Photo by YAROSLAV SAPRYKIN on Unsplash
Why is it important?
This paper summarizes the 3D targeting methods and analyzes their performance and methods, analyzes the improvement measures as well as the future development trends.
Perspectives
As the development of autonomous driving technology progresses, the requirements for data sets are constantly increasing. Finding suitable and efficient automatic annotation methods is one of the key factors in the development of autonomous driving technology. This paper summarizes a series of methods and analyzes their advantages and disadvantages, which can provide reference for future technological development.
航 尹
Guangdong University of Technology
Read the Original
This page is a summary of: Automated driving: Study on 3D target automation labeling, January 2024, American Institute of Physics,
DOI: 10.1063/5.0222873.
You can read the full text:
Contributors
The following have contributed to this page
