The SIVAL (Spatially Independent, Variable Area, and Lighting) benchmark
SIVAL includes 25 different image categories with 60 images for each
category. This benchmark
emphasizes the task of Localized CBIR. The categories consist of
images of single objects photographed against highly diverse
backgrounds. The objects may occur anywhere spatially in the image
and also may be photographed at a wide-angle or close up. We have
created this benchmark since most of the Corel object images contain
close-ups of an object that is centered in the image and typically
occupies a majority of the image.
All pictures were taken with a Cannon s230 Digital Camera from a tripod
at a resolution of 1024 by 768. For all pictures the flash was disaples,
the JPEG setting was super-fine, the ISO speed was automatic and the white
balance was on a manual setting.
Zip file with the unprocessed images
Zip file with
the bags. This zip file includes the representation of the SIVAL
images as bags used by Accio (for IHS with 32 segments when using neighbors),
thumbnails for the segmented images, java code to read the files
holding the bags, and documentation on the file format.