SEEMORE DATABASE
is the database of images used to train and test the system called SEEMORE, described in
- Mel, B.W. (1997) SEEMORE: Combining color, shape, and texture histogramming in a neurally-inspired approach to visual object recognition. Neural Computation, 9, 777-804. [PDF]

The database consists of color video images of a set of views of 100 real objects or photographs of complex scenes including objects which are rigid, non-rigid, simple, complex, bland, colorful, etc. Thumbnail photos of the 100 objects are available on this page. Important information about the database, its structure, and its allowable uses can be found here.

The list of 100 objects can be found here, while the pictures themselves can be downloaded as a single 27Mb tar file, or individually.

The code is all available here with NO SUPPORT WHATSOEVER. All questions about this database and its documentation should be addressed to Bartlett Mel at mel@usc.edu.