Video Tracking Database (Human)
Overview
A lot of algorithms for object tracking were proposed in the past from different groups. To experimentally show good performance of an algorithm every group used their own set of videos. This Video Tracking Database was designed to compare tracking algorithms on natural human tasks. The Video Tracking Database provides an extensive selction of different tasks and motions found in real life. Additonally to providing plain videos, data for three trackers (nearest neighbour tracker, Baker+Matthews inverse compositional tracker and efficient second order minimization tracker) running on the videos is supplied.
How to use the Video Tracking Database
This database contains several videos to test trackers. There are oriented motion videos (object motion parralel to picture plain, motion around object axis and rotation around object axis) as well as complex task videos available. Tasks come in five different speeds plus one with increasing speed during motion. The videos were recorded in two light conditions: normal light (unmodified office light) and diffuse light (screen kept out direct light, noise occurs) with less reflections on the objects.
All available videos were recorded with 30 fps. On the following page you can find the videos organized in tables. A thumbnail image shows which task to expect and some text gives information about the main challenge for the tracker and how the trackers perform. On which frame a tracker fails can be red in the .txt file after clicking "see details". Following "downloads" you can choose from downloading .jpg files and a tracking data file of the speed and light condition you need.
| nnbmic |
nearest neighbour tracker |
T. Dick, C. Perez, M. Jagersand, A. Shademan. Realtime Registration-Based Tracking via Approximate Nearest Neighbour Search. Proceedings of Robotics: Science and Systems, 2013 |
| bmic |
SSD/Baker+Matthews inverse compositional tracker | S. Baker and I. Matthews.
Equivalence and efficiency of image alignment algorithms. Computer
Vision and Pattern Recognition, 2001. CVPR 2001. Proceedings of the
2001 IEEE Computer Society Conference on, 2001 |
| esm |
efficient second order
minimization tracker |
S. Benhimane, E. Malis.
Real-time image-based tracking of planes using Efficient Second-order
Minimization. Intelligent Robots and Systems, 2004. (IROS 2004)
Proceedings of 2004 IEEE/RSJ International Conference on |
A writeup with technical information about the Video Tracking Database can be found here: PDF
Single Motion Videos
Section of the actual page you can find after following the link:
Composite Motion task videos
Section of the actual page you can find after following the link: