VIDEO OBJECTS

-- a test database for video object recognition --

purpose:

This dataset provides small 180 videos showing 15 objects. It serves as a testbed for the training and recognition of object appearance in video streams, with the focus on a real-world object recognition scenario including clutter, motion blur, and illumination changes.

download:

Download the data here (30.3 MB, video_objects.tar.gz)

generation:

the videos were generated by choosing 15 objects and manually presenting them to a web-cam (UniBrain Fire-I) at a framerate of 25/s. For each object, 12 videos were taken, each providing about 40 frames at a 320x240 resolution.

characteristics:

While the objects and the arm of the operator move, the background remains static. The videos were taken at two different locations providing a different background and different illumination:

1) office: frames are a generally darker. and the background is monotonous except for a few strong edges.

2)lab: the scenes contain strong daylight and many specular highlights. The background is strongly textured.

though a front side is chosen for each object, slight pose changes occur due to the manual presentation. Also, a strong motion blur can be observed in many frames.

publications:

Adrian Ulges, Daniel Keysers, Christoph Lampert, Thomas Breuel: Improving Object Recognition in Videos using Motion-based Segmentation. 28th Annual Symposium of the German Association for Pattern Recognition, 2006. submitted for review.

contact:

IUPR Research Group

TU Kaiserslautern / DFKI

ulges at iupr.net