Research Datasets

This page contains datasets acquired by CRS4 Visual Computing for research purposes. The available datasets may be downloaded and used for research purposes by acknowledging CRS4 for the data acquisition in any resulting publication or report as indicated below.

360-degree captures of indoor environments

Dataset	Description	Data
	Apartment0 - 360 degree capture of a multi-room residential environment (D1 - 11 rooms).	36 equirectangular images (360x180 degrees, 5376x2688) in JPEG format. First camera height: 170 cm. [Download zip], 134 MB.
	Apartment1 - 360 degree capture of a multi-room residential environment (D2 - 5 rooms).	17 equirectangular images (360x180 degrees, 5376x2688) in JPEG format. First camera height: 170 cm. [Download zip], 69 MB.
	Loft - 360 degree capture of a double sloped ceiling loft (D3 - 1 room).	17 equirectangular images (360x180 degrees, 5376x2688) in JPEG format. First camera height: 110 cm. [Download zip], 27 MB.
	Attic - 360 degree capture of an attic with sloped ceiling (D3 - 1 room).	8 equirectangular images (360x180 degrees, 5376x2688) in JPEG format. First camera height: 110 cm. [Download zip], 30 MB.
	Office0 - 360 degree capture of a multi-space commercial environment.	49 equirectangular images (360x180 degrees, 5376x2688) in JPEG format. First camera height: 170 cm. [Download zip], 188 MB.
	Office1 - 360 degree capture of a cluttered commercial environment (same of Office0 with clutter).	47 equirectangular images (360x180 degrees, 5376x2688) in JPEG format. First camera height: 170 cm. [Download zip], 180 MB.
	Cellar - 360 degree capture of a large cellar (D5 - 2 rooms).	17 equirectangular images (360x180 degrees, 5376x2688) in JPEG format. First camera height: 160 cm. [Download zip], 43 MB.

Reconstructions of indoor environments

Dataset	Description	Data
	Apartment0 reconstruction and rectified images - Results associated with the paper: Recovering 3D existing-conditions of indoor structures from spherical images. G. Pintore, R. Pintus, F. Ganovelli, R. Scopigno, E Gobbetti.	Obj files and rectified images for object detection. [Download zip], 36 MB.
	Office0 reconstruction and rectified images - Results associated with the paper: Recovering 3D existing-conditions of indoor structures from spherical images. G. Pintore, R. Pintus, F. Ganovelli, R. Scopigno, E Gobbetti.	Obj files and rectified images for object detection. [Download zip], 20 MB.

Datasets made available for research purposes by CRS4 Visual Computing