The USyd Campus Dataset

Citation Author(s):
Wei
Zhou
The University of Sydney
Julie Stephany
Berrio Perez
The University of Sydney
Charika
De Alvis
The University of Sydney
Mao
Shan
The University of Sydney
Stewart
Worrall
The University of Sydney
James
Ward
The University of Sydney
Eduardo
Nebot
The University of Sydney
Submitted by:
Wei Zhou
Last updated:
Sun, 10/13/2019 - 19:40
DOI:
10.21227/sk74-7419
Data Format:
Links:
License:
Creative Commons Attribution
1599 Views
Categories:
Category: 
Keywords:
0
0 ratings - Please login to submit your rating.

Abstract 

Vision and lidar are complementary sensors that are incorporated into many applications of intelligent transportation systems. These sensors have been used to great effect in research related to perception, navigation and deep-learning applications. Despite this success, the validation of algorithm robustness has recently been recognised as a major challenge for the massive deployment of these new technologies. It is well known that algorithms and models trained or tested with a particular dataset tend not to generalise well for other scenarios. For certain applications, autonomous vehicles in particular, algorithm failures can lead to potentially catastrophic consequences. To minimise this risk, it is essential to evaluate the robustness and reliability of algorithms before deploying them in real life applications. This paper presents a long-term, large-scale dataset collected over the period of 1.5 years on a weekly basis over the University of Sydney campus and surrounds. The USyd Campus Dataset has been designed to aid in the development, validation and testing of algorithm robustness under a wide variety of conditions. It includes multiple sensor modalities and covers various environmental conditions as well as diverse changes to illumination, scene structure, and pedestrian/vehicle traffic volumes. The dataset also includes a semantic segmentation dataset with different illumination conditions and camera perspectives, and a ‘Road’ class image dataset which is automatically generated using lidar. To facilitate the dataset use by researchers and practitioners, we also provide a set of development tools for the analysis and visualisation of the dataset.

Instructions: 

For detailed information about this dataset and the tools, please go to our website: http://its.acfr.usyd.edu.au/datasets/usyd-campus-dataset/

 

Dataset Files

ACCESS ON AWS