Hide menu

PAnorama Sparsely STructured Areas Datasets

We present the PASSTA datasets used for evaluation of the image alignment methods in the paper: The images are assumed to have been taken from a camera fixed on a tripod. Between each subsequent image, the camera is assumed to have been rotated around the vertical axis through the optical center. The rotations are further assumed to be small enough such that two subsequent images have a significantly overlapping view of the scene, at least 50%. We assume known camera intrinsic parameter and the lens distortions (the calibration parameter are included in the respective datasets).
  • Synthetic : consists of 72 images of a room rendered with Blender with a resolution of 1280x1920 px. Intrinsic parameters were retrieved from Blender and the camera is rotating by exactly 5 degrees between consecutive images. This dataset depicts an almost sparsely structured scene.
    download (74 MB)
  • Lunch Room Blue: consists of 72 images acquired with a Canon DS50 and perspective lenses with a resolution of 1280x1920 px at poor light condition. A panorama head was used to approximate a fixed rotation of 5 degrees around the vertical axis about the optical center of the camera.
    download (278 MB)
  • Lunch Room: consists of 72 images acquired with a Canon DS70 and wide angle lenses Samyang 2.8/10mm (about 105 degree), with a resolution of 5740x3780 px. A panorama head was used to approximate a fixed rotation of 5 degrees around the vertical axis about the optical center of the camera.
    download (63 MB)

Acknowledgements:
The authors thank Andreas Robinson for providing the synthetic dataset.


Last updated: 2015-07-10