A Survey on Autonomous Driving Datasets

Currently the autonomous driving is one of the rapidly developing and widely concerned fields, which involves numerous artificial intelligence algorithms. Generally, these artificial intelligence algorithms require a large amount of diverse data for training and testing to ensure their performance i...

Full description

Saved in:

Bibliographic Details
Published in:	2021 8th International Conference on Dependable Systems and Their Applications (DSA) pp. 399 - 407
Main Authors:	Liu, Weiyu, Dong, Qian, Wang, Pengqi, Yang, Guang, Meng, Lingzhong, Song, You, Shi, Yuan, Xue, Yunzhi
Format:	Conference Proceeding
Language:	English
Published:	IEEE 01-08-2021
Subjects:	Adaptation models Automatic labeling Autonomous driving Data models Data type Dataset Domain adaptation Labeling Laser radar Millimeter wave radar Radar imaging Training
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Currently the autonomous driving is one of the rapidly developing and widely concerned fields, which involves numerous artificial intelligence algorithms. Generally, these artificial intelligence algorithms require a large amount of diverse data for training and testing to ensure their performance in actual applications. In this paper, we introduce the types of data related to the autonomous driving, compare multiple datasets, and discusse the key issues in the process of dataset collection and generation.
ISSN:	2767-6684
DOI:	10.1109/DSA52907.2021.00060