UNITN Crowd Dataset Descriptions:

The UCD Dataset consists of two video sequences which have been segmented into two different sub-sequences each, and used for both crowd motion segmentation and anomaly detection.

All the videos are taken outdoors in '.avi' format, with the resolution of 320*240, frame rate 30fps.

Segment 1; Segment 2;
Segment 3; Segment 4;
Video sequences are provided in the .rar file:
Version 1.0 - Videos Sequences
Please cite this dataset as:
     title={Crowd motion segmentation and anomaly detection via multi-label optimization},
     author={Ullah, Habib and Conci, Nicola},
    journal={ICPR workshop on Pattern Recognition and Crowd Analysis},
All rights reserved. Multimedia Signal processing and Understanding Lab. - University of Trento