The Drive&Act dataset is a state of the art multi modal benchmark for action recognition in automated vehicles. It offers following key features:
Copyright Fraunhofer IOSB.
Usage for research only.
To gain access to the dataset please use the contact form or write to both Manuel Martin and Alina Roitberg.
|A-Column co-driver||NIR Depth RGB Kinect IR|
|Annotations||Activities 3D Body Pose Interior|
If you use the dataset please cite the following publication.
Configured the webpage. Providing the data used in the ICCV publication.
Coming soon: head pose, continuous labels and a more detailed interior model