TrakMark

Last Update : June 28, 2010

NAIST Campus Package 01

Abstract

This package contains two kinds of image sequences, one taken with a standard monocular camera and one taken with an omnidirectional multi-camera type. Image sequences were taken in the outdoor environment where there are both natural and artificial objects. For both kinds of sequences, calibration parameters and estimated camera paths were prepared. In simple use, a tracking method can be tested using one of these sequences. In more exhaustive use, the monocular and omnidirectional sequences can be used for testing and learning, respectively. For learning, 2D and 3D positions of feature points calculated by a structure-from-motion method were also prepared. For various comparisons, all of the camera paths and feature point data were described in the same coordinate system.

Image Sequence

Information

Naist Campus Package 01
monocular camera omnidirectional camera
Frame Rate 15 (FPS) 15 (FPS)
Image Resolution 720 x 480 768 x 1024 (x 6 internal cameras)
File Size About 200KB per frame About 250KB per frame
File Format JPEG JPEG
Loss of Frame None None
Reference Point Exist Exist
Number of Sequences 1 1 (x 6 internal cameras)
Creator Nara Institute of Science and Technology
Last Update 2010/06/28

[Text File : history.txt]Update History

Image Sequence Files

Mechanical Equipment

Mechanical Equipment of Naist Campus Package 01.
Monocular CameraSony DSR-PD150
Lens of Monocular CameraSony VCL-HG0758
Omnidirectional CameraPoint Grey Research Ladybug (version 1)

Shooting Environment

Snapshot.
AerialImage

Result of Camera Calibration

Reference

  1. R. Y. Tsai: "An efficient and accurate camera calibration technique for 3D machine vision," Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 364–374, 1986.
  2. T. Taketomi, T. Sato, and N. Yokoya: "Real-time geometric registration using feature landmark database for augmented reality applications", Proc. SPIE Electronic Imaging, Vol. 7238, 2009.
  3. S. Ikeda, T. Sato, and N. Yokoya: "Panoramic movie generation using an omnidirectional multi-camera system for telepresence", Proc. 13th Scandinavian Conf. on Image Analysis, pp. 1074-1081, 2003.
  4. T. Sato, S. Ikeda, and N. Yokoya: "Extrinsic camera parameter recovery from multiple image sequences captured by an omni-directional multi-camera system", Proc. European Conf. on Computer Vision, Vol. 2, pp. 326-340, 2004.

Warning

Please note that these video clips were provided just to exemplify each scenario, and do not follow the video format guidelines.