Abstract: 3D vision of image has become the area of interest of many researchers. Numerous image processing algorithms have been developed over the past decade. This work covers a novel method of converting 2D image to 3D image. 3D imaging provides very detailed and valuable information. The approach effectively incorporates visually prominent boundaries, prominence measure of pixels and local color & region gestures. The proposed algorithm uses the method of gradient that would be the direction of uphill. The gradient has a direction and a magnitude, it is natural to encode this information in a vector. The length of this vector provides the magnitude of the gradient. Vector can be used to provide height/depth of an image, that generates 3D image of an original 2D digital image.
Keywords: 3D, 2D, reconstruction, gradient, displacement map,