Estimation of Depth Information from a Single View in an Image

Distance measurement in real world has always been one of the challenging tasks to be performed in the field of computer vision. Photographic images are twodimensional depiction of three-dimensional real space. On methodical observation of the same, one can identify some third dimensional properties (like depth information). In this paper we have proposed a method, which employs perspective projective geometrical tool, and henceforth bring out a new idea of computing the distances between the edges in the object, which are parallel to the image plane in photographs of actual scene. The technique presented employs uncalibrated images with no knowledge of the internal parameters of the camera (as focal length & aspect ratio) or it’s pose (position & orientation with respect to viewed scene). Geometric characteristics (perspective projections) of the scene are employed.