Published March 29, 2022 | Version v1
Dataset Open

VWFP: Virtual World Fallen People Dataset for Visual Fallen People Detection

  • 1. Institute of Information Science and Technologies (ISTI-CNR), Pisa, Italy
  • 2. University of Pisa, Italy

Contributors

Data collector:

Data curator:

  • 1. Institute of Information Science and Technologies (ISTI-CNR), Pisa, Italy
  • 2. University of Pisa, Italy

Description

A synthetic dataset for visual fallen people detection comprising images extracted from the highly photo-realistic video game Grand Theft Auto V developed by Rockstar North. Each image is labeled by the game engine providing bounding boxes and statuses (fallen or non-fallen) of people present in the scene.
The dataset comprises 6,071 synthetic images depicting 7,456 fallen and 26,125 non-fallen pedestrian instances in various looks, camera positions, background scenes, lightning, and occlusion conditions.

Files

vwfp-dataset.zip

Files (11.5 GB)

Name Size Download all
md5:359155cfb3b2bb36db83a5fce136d56b
11.5 GB Preview Download

Additional details

Funding

AI4Media – A European Excellence Centre for Media, Society and Democracy 951911
European Commission