• About
  • About the Department
  • Contact
  • Directions
  • Partners
  • Rooms
  • Systems & Software
  • Conduit
  • News
  • Events
  • Blog
• People
  • Our Community
  • Faculty
  • Staff
  • Grad Students
  • Ugrad Students
  • Alums
  • Directory
• Research
  • Research Links
  • Areas
  • Publications
• Degrees
  • Degree Programs
  • Doctoral
  • Masters
  • Undergrad
  • Programs
  • Miscellaneous
• Courses
  • Course List
  • Schedule
  • Sections
  • TA Program

The 27th IPP Symposium

The Open Source Computer Vision Library--New Developments

Gary R. Bradski, Intel Corporation

The Open Source Computer Vision Library (OpenCV) is a collection of C and light C++ routines running under Linux and Windows intended to support research and to help create a market in consumer computer vision. As such, the source is open and is free for research or commercial use under a BSD- style license (user's source need not be open). An active user group of over 1500 users may be joined by sending email to OpenCV-Subscribe@yahoogroups.com. We have recently posted OpenCV to SourceForge to allow greater community contributions.

I will overview the library; demonstrate automatic camera calibration, embedded-HMM-based face recognition, and real- time performance analysis. I will end with new results in extremely efficient and lighting stable image segmentation and demonstrate its use in background differencing.