Research Data Leeds Repository

MINDSEYE2015: Joint Tracking and Event Analysis for Carried Object Detection.

Tavanai, Aryana (2015) MINDSEYE2015: Joint Tracking and Event Analysis for Carried Object Detection. University of Leeds. [Dataset]

Dataset description

MINDSEYE2015 is a dataset of videos showing a variety of people interacting with various different objects. It was primarily created to improve carried object tracking using event analysis. It consists of 15 videos (5 recordings captured from 3 viewpoints), each lasting approximately 6000 frames. These videos have a resolution of 360x640 at 20 frames per second and are taken from three different viewpoints which allows for a better evaluation in dealing with object occlusions. In the majority of frames there are at most one person and one object, but there are cases of more or less than one person or object present in the scene. Since the dataset was captured outdoors, the movement of trees and cloths on the table, as well as the change in brightness of the video due to clouds and distance of the person to the camera cause challenges for object detectors. We have defined 7 events, namely Carry, Static, Pickup, Putdown, Drop, Raise and Roll, to allow for a full description of the scene with regards to the state of the carried object, from the start of its appearance to its disappearance. Ground truth for person tracks, carried object tracks and events are fully annotated. Carried object detections, tracklets and tracks are also provided. For more information please visit:

Keywords: tracking, carried object detection, event analysis
Subjects: I000 - Computer sciences > I400 - Artificial intelligence
Divisions: Faculty of Engineering and Physical Sciences > School of Computing
Related resources:
License: Creative Commons Attribution 4.0 International (CC BY 4.0)
Date deposited: 28 Jul 2015 07:37




Research Data Leeds Repository is powered by EPrints
Copyright © 2024 University of Leeds