"GrOVis-Fire"
A multi-video sequence for volumetric reconstruction and rendering research

Introduction
Download
Calibration How-To
Additional notes and news
Contact information
Other multi-view video sequences

Introduction

This is the web-home for the GrOVis-Fire multi-video data set. The data is meant to provide real world test data for sparse view volumetric reconstruction algorithms.

The set consists of frame sequences recorded from 8 different camera views located in a ring around the scene and the corresponding camera parameters. The recorded scene contains a burning alcohol flame on black background. The sequences are 346 frames long and each frame is saved as a 320x240 image. Some snapshots from the sequences can be seen in Figure 1.

Figure 1. One flame seen from different views.

Download

Please be advised that the file is quite large.

The Data set

File	Description	Size
grovis_fire.tar.gz	8 view fire sequence + calibration data	204 MB

Camera Calibration How-To

The calibration data is stored in a XML-File ( GrOVis_Fire/tmp/ec.xml after unpacking the sequence ) it contains Hartlet/Zisserman style calibration matrices ( K - internal calibration matrix, R - rotation matrix, t - translation vector ) + one additional "flip matrix" which is used to convert left-handed calibrations to right-handed coordinate systems and vice versa.

The matrices are stored the following way:

int m
int n
float a11 a12 ... a1n a21 a22 ... a2n ... am1 am2 ... amn

The projection matrix P for a right-handed coordinate system is computed by

P = K * M * R * [ 1 | -t ]

If you are using a left-handed coordinate system try setting M to

[ -1  0  0 ] 

[  0  1  0 ] 

[  0  0  1 ] 

Notes and news

• If you note any error in this data or have additional information that would help other people to use the data please let us know (ihrke@mpi-sb.mpg.de).

Contact Information

If you have any questions or comments you are welcome to send them to ihrke@mpi-sb.mpg.de.

www.grovis.de
Graphics-Optics-Vision
Max-Planck-Institut für Informatik
Stuhlsatzenhausweg 85
D-66 123 Saarbrücken
Germany

Other multi-view video sequences