Welcome to the SaliencyToolbox!

If you are in a hurry to get started, then jump to the download instructions right away!

Changes:

• Fix of a bug in the iterative normalization procedure that caused spurious saliency activity at the edges and, in particular, in the corners of feature maps and conspicuity maps.
• Empty maps are now displayed as black instead of gray.
• A new top-level function batchSaliency.m was added for easy batch processing of multiple images.

SaliencyToolbox 2.0 is finally here!

New features in version 2.0:

• Graphical User Interface for intuitive control of parameters
• Lighter code for faster execution
• Visualization mode with contrast modulation
• Support for Intel Macs
• New license agreement

The SaliencyToolbox is a collection of Matlab functions and scripts for computing the saliency map for an image, for determining the extent of a proto-object, and for serially scanning the image with the focus of attention.

System requirements:

• Any computer and operating system that runs Matlab
• Matlab Release 13 or later
• Image Processing Toolbox

The toolbox contains pre-compiled binary mex files for MS Windows, Mac OS X (both Power PC and Intel Macs), and Linux (32 bit and 64 bit). The source code can be compiled on any system with the GNU C compiler gcc.

The SaliencyToolbox is in part a reimplementation of the iNVT toolkit at Laurent Itti's lab at the USC. This toolbox complements the iNVT code in that it is more compact (about 5,000 versus 360,000 lines of code) and easier to understand and experiment with, but it only contains the core functionality for attending to salient image regions.

Although time critical procedures are contained in mex files, processing an image with the SaliencyToolbox in Matlab takes longer than with the iNVT code. Whenever processing speed or feature richness is paramount, the iNVT code should be preferred. For computing the saliency map or attending to salient proto-objects in an image in a transparent and platform independent way, the SaliencyToolbox is a good choice.

The code was originally developed as part of Dirk B. Walther's PhD thesis in the Koch Lab at the California Institute of Technology, made possible by funding from NSF and NIMH.