When looking at a photograph, humans usually have no trouble understanding the 2-dimensional image as a reproduction of a 3-dimensional world. Yet one of the difficulties in computer vision is making a computer understand that as well. Much research has been done to find techiniques for accomplishing this task, and much needs still to be done. The following information will hopefully give you some insight into a few different techiniques which have been developed.

• Line Labeling
  • Types of Junctions and Lines
  • Impossible Shapes
• Binocular Stereopsis
  • Disparity
  • Conjugate Pairs
  • The Epipolar Line
• Motion
  • Binary Difference Pictures
  • Motion Correspondence
    • Relaxation Labeling
  • Image Flow
    • Focus of Expansion
  • No Hands Across America
• Focus