OmniScape

The OmniScape dataset consists of 10,000 annotated images captured from four different view angles: Front View (FV), Mirror-View Right (MVR), Mirror-View Left (MVL), and Rear View (RV) of a vehicle. The dataset provides the semantic annotations for ten classes, including road, lane markings, curb, person, rider, car, bicycle, motorcycle, traffic sign, and background.

Data and Resources

Cite this as

OmniScape (2024). Dataset: OmniScape. https://doi.org/10.57702/vhcel44s

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.48550/arXiv.2407.16647
Author OmniScape
Homepage https://doi.org/10.1101/2020.05.01.2009341