Abstract:
In each right and left frame of a stereo image, a maximum disparity value corresponding to a foremost object is measured through a simple block matching method. A point where the maximum disparity value or a color histogram is suddenly changed is estimated at a point where three-dimensional scene change is performed in a flow of a continuous stereo frame. When the maximum disparity value is excessively large in the frame where the scene change is performed, it may cause visual discomfort so that a horizontal parallax is controlled to reduce the excessive parallax.
Abstract:
Provided is an image processing device which includes a first calculation part which calculates a first position of at least one first point sampled from a real 3D object when a 3D image is acquired, a second calculation part which calculates a second position of at least one second point provided from a reception part by corresponding to the first point, by using at least one second parameter related to the reception part provided from the 3D image, and a decision part which decides at least one first parameter related to a transmission part to be provided to the reception part, by acquiring the 3D image to minimize a difference between the first position and the second position.
Abstract:
PURPOSE: An apparatus and a method for efficient viewer-centric depth adjustment based on virtual front-parallel planar projection in stereoscopic images are provided to rapidly and effectively adjust a depth of an image while not allocating resources to calculation of an accurate binocular disparity map. CONSTITUTION: A depth adjusting unit (120) sets a second point by moving the coordinates of a first point of a first image in a viewer centric direction to correspond to a depth adjustment value. The depth adjusting unit sets a third point by projecting the second point in a direction heading for the second point from the user viewpoint associated with the first image. An image synthesizing unit (130) determines color values of the third point and generates a second image by adjusting the depth of at least a partial region of the first image. [Reference numerals] (110) Plane determining unit; (120) Depth adjusting unit; (130) Image synthesizing unit