Learning-Based Computer Vision with OpenCV

TitleLearning-Based Computer Vision with OpenCV
Publication TypeJournal Article
Year of Publication2005
AuthorsBradski, Gary., Kaehler, Adrian., and Pisarevsky, Vadim
JournalIntel Technology Journal
AbstractThe rapid expansion of computer processing power combined with the rapid development of digital camera capability has resulted in equally rapid advances in computer vision capability and use. Intel has long been at the forefront of enabling this advance on the computer hardware and software side. Computer vision software is supported by the free Open Source Computer Vision Library (OpenCV) that optionally may be highly optimized by loading the commercial Intel Integrated Performance Primitives (IPP). IPP now automatically supports OpenCV with no need to change or even recompile the user’s source code. This functionality enables development groups to deploy vision and provides basic infrastructure to experts in vision. OpenCV has long supported “geometric vision” from camera calibration, motion tracking in 2D, finding the camera location given a known 3D object, on up to producing depth maps from stereo vision. This paper describes using OpenCV for “learning-based vision,” where objects such as faces, or patterns such as roads, are learned for segmentation and recognition.
vol09_art03.pdf460.58 KB