Welcome to the Open Source Computer Vision Library. Gigahertz microprocessors, falling camera prices and ten times more video capture bandwidth from USB2 are providing new opportunities to apply computer vision anywhere you might find a PC. Real time gestural control around your home PC is now conceivable. The Visual Interactivity Group in Intel's Microprocessor Research Lab initiated the development of this library 2 years ago and has made two alpha releases of this code prior to this first Open Source release. Our goal is to establish an open source vision community and provide a site where the distributed efforts of the community can be consolidated and performance optimized. The library is intended for use by researchers and commercial software developers. Computer vision research is filled with many competing groups and ideas with no unifying theory of vision, typifying Thomas Kuhn's definition of a "pre-paradigmatic science". In Kuhn's Structure of Scientific Revolutions, he argues that progress in a given scientific field derives from the acceptance of a common paradigm by which members of the group can share and compare. We cannot offer here that unifying theory of vision by which the field will cohere, but we can offer the community a central infrastructure by which we can share and compare. This is the driving research philosophy of this library, and by this we hope at least technical, if not scientific progress can be made. To involve and benefit from the combined expertise of the vision community we have recruited a committee of experts to participate in acceptance decisions for new content for the library and in setting the library direction in general.  Intel's contribution is to provide its own research contributions, host the library on a website, and contribute assembly language optimized versions of the most compute intensive code. We thank all the current contributors to the library and encourage others to join by submitting code. The Release Notes provide information about system requirements, directory contents, installation notes, frequently asked questions, support, and where to send bug reports and suggestions. The Open Source License for the library describes use and redistribution rights for the library.

Function Areas	Application Areas
These are the general function areas currently supported in the library. Clicking on a function area will bring up another browser window that discuses each area in more detail. Click here for full function documentation in PDF format. Geometric Methods Recognition Measures Segmentation Utilities Features Image Pyramids Camera Tracking Fitting Matrix Image Processing	We have included sample applications (and source code!) that use the library. Clicking on an application will open another browser window that discusses the each application in more detail. Camera Calibration Tool Face Recognition Tool   Pyramidal Lucas Kanade Feature Tracker CAMSHIFT Tracker   Kalman Tracker Condensation Tracker

Open Source Computer Vision Library Committee

Back to Top