SIFT: Scale Invariant Feature Transform

3D Image Recognition Technology for Machine Vision


  • Robust and well-proven method for object recognition
  • Recognize objects with partial visibility, different lighting conditions or zoom
  • Extends object recognition to 3D

Technology Details

Researchers at The University of British Columbia have developed an advanced technology that allows computers to recognize objects in digital images.  Given a few previous images of an object, the system can recognize the object in new images even under different lighting conditions, magnification/zoom, and 3-D orientations, including when it is partially hidden in a scene.

The technology has proven effective in a large range of applications to detect local features in images. SIFT is widely recognized as a ‘best in class’ solution for image recognition and detection. The technology is patented and has been licensed by many companies for use in a wide variety of products and markets.


Non-exclusive licenses to the patent are available in various fields of use. Parties interested in licensing SIFT should direct queries to the listed Contact for further information.