User Tools

Site Tools


people:gossow:rgbd

This is an old revision of the document!


RGBD Datasets

These data sets resemble the structure of the datasets created by Mikolajczyk et. al., however they contain additional depth maps and information about the 3D camera movement.

All datasets were captured using a Kinect camera. Each contains a series of RGB and depth images with a transformation of increasing strength.

Data description

Each directory contains the following files:

  • depth*.pgm: 16-Bit depth images in mm
  • img*.pgm: 8-Bit RGB images
  • mask.pgm: Indicates region of interest in first image
  • H1to*p: 3×3 Homography from image 1 to n
  • K: 3×3 intrinsic camera matrix
  • cam_rotation: rotation around camera axis for all images (except first)
  • vp_rotation: rotation around plane normal
  • vp_angle: angle between camera axis and plane normal (relative to img1)
  • scaling: 2d image scaling relative to first image

The 'Poster' Data Set

Download the full dataset: poster.tar.gz (623 MB)

This scene shows several posters fixed to the ground or wall. It contains the following transformations:

  • camrotate,camrotate45,camrotate60: The camera rotates up to 90° around its view axis, while looking at the posters in a fronto-parallel view or 45°/60° tilted to the side.

  • scale: The camera moves away from the scene, resulting in a scaling effect down to 24%.

  • vpangle, vpangle22, vpangle45: The camera starts at a fronto-parallel view, tilting to the side relative to the plane normal. The optical tilt axis is at 0, 22 and 45 degrees.

  • vprotate35,vprotate40,vprotate45,vprotate55: The camera is at 35°/40°/45°/55° relative to the plane normal and rotates around that normal.

people/gossow/rgbd.1341413467.txt.gz · Last modified: 2012/07/04 16:51 by gossow