Abstract: Provided are a method for image prediction of a multi-view video codec capable of improving coding efficiency, and a computer readable recording medium therefor. The method for image prediction of a multi-view video codec includes partitioning an image to a plurality of base blocks, acquiring information of reference images which are temporally different, acquiring information of reference images which have different views, and predicting a target block based on the acquired information. Accordingly, an image that is most similar to an image of a view to be currently compressed is generated using multiple images of different views, so that coding efficiency can be improved.

Abstract: An image synthesizing method according to the present invention includes generating a depth image in a current picture by searching depth information using a plurality of color images obtained at different view points, performing filtering on the depth image using a 3-dimensional (3D) joint bilateral filter, and generating a synthesized image using the plurality of color images and the filtered depth image, wherein the 3D joint bilateral filter performs filtering on the generated depth image using color image information for at least one of previous pictures, the current picture, and subsequent pictures, and the color image information includes information on a boundary of an object in the color images and color information of the color images. According to the present invention, image processing performance may be enhanced.

Abstract: Method for filling a hole-region and a three-dimensional (3D) video system using the same. In order to synthesize an image, a plurality of images each having a different viewpoint may be obtained. By using depth maps of the obtained plurality of images, an intermediate virtual viewpoint image may be synthesized. After synthesizing, a hole-region in the intermediate virtual viewpoint image may be filled using a color image of the intermediate virtual viewpoint image, a mask image indicating the hole-region in the intermediate virtual viewpoint image, and a depth image of the intermediate virtual viewpoint image.

Abstract: Provided are a method for image prediction of a multi-view video codec capable of improving coding efficiency, and a computer readable recording medium therefor. The method for image prediction of a multi-view video codec includes partitioning an image to a plurality of base blocks, acquiring information of reference images which are temporally different, acquiring information of reference images which have different views, and predicting a target block based on the acquired information. Accordingly, an image that is most similar to an image of a view to be currently compressed is generated using multiple images of different views, so that coding efficiency can be improved.

Abstract: In a method for image prediction of a multi-view video codec, a disparity vector value of a current macroblock is obtained and a corresponding macroblock is searched through the obtained disparity vector value. An intra prediction coding direction value of the searched macroblock is determined as an intra prediction coding direction value of the current macroblock. The current macroblock is coded/decoded through the determined intra prediction coding direction value. The coding efficiency can be increased by reducing a coding amount by omitting a process of obtaining an unnecessary disparity value by using a different-view intra prediction coding direction as a current-view intra prediction coding direction. The coding efficiency can be increased by reducing a coding amount by differentially coding a similar disparity vector. The reliability can be increased through an accurate disparity using multiples of 8 rather than 16, in motion skip for sharing the format of a macroblock.

Abstract: An image synthesizing method according to the present invention includes generating a depth image in a current picture by searching depth information using a plurality of color images obtained at different view points, performing filtering on the depth image using a 3-dimensional (3D) joint bilateral filter, and generating a synthesized image using the plurality of color images and the filtered depth image, wherein the 3D joint bilateral filter performs filtering on the generated depth image using color image information for at least one of previous pictures, the current picture, and subsequent pictures, and the color image information includes information on a boundary of an object in the color images and color information of the color images. According to the present invention, image processing performance may be enhanced.

Abstract: The present invention relates to a device and a method for estimating a depth map, and a method for making an intermediate image and a method for encoding multi-view video using the same.

Abstract: Method for filling a hole-region and a three-dimensional (3D) video system using the same. In order to synthesize an image, a plurality of images each having a different viewpoint may be obtained. By using depth maps of the obtained plurality of images, an intermediate virtual viewpoint image may be synthesized. After synthesizing, a hole-region in the intermediate virtual viewpoint image may be filled using a color image of the intermediate virtual viewpoint image, a mask image indicating the hole-region in the intermediate virtual viewpoint image, and a depth image of the intermediate virtual viewpoint image.

Abstract: The invention provides a method of encoding a multi-view video and an encoder capable of calculating global disparity between pictures at different viewpoints in the multi-view video and forming a hierarchical B picture in both view and time directions on the basis of the calculated global disparity, thereby improving encoding efficiency. According to an embodiment of the invention, there is provided an encoder for encoding data for pictures captured at different viewpoints in a multi-view video. When a picture at one of the first and last viewpoints is encoded into an I picture, the encoder encodes a picture that is captured at the other viewpoint but is disposed in the same time direction into an VP picture.

Abstract: The present invention relates to a method and apparatus for encoding a multi-view video and a method and apparatus for decoding a multi-view video, and more particularly, to a method and apparatus for encoding a multi-view video and a method and apparatus for decoding a multi-view video that can perform encoding and decoding in consideration of global disparity between pictures captured at two different viewpoints to remove spatial redundancy. An embodiment of the invention provides an encoder for compressing data for pictures captured at different viewpoints using global disparity in a multi-view video. The encoder includes: an input unit that receives pictures captured at two or more viewpoints; a reference picture generating unit that generates a new reference picture on the basis of the global disparity between a target picture and a reference picture; and a motion predicting unit that calculates motion vectors of the target picture on the basis of the reference picture.

Abstract: The invention relates to a method of computing a disparity, a method of synthesizing an interpolation view, a method of encoding and decoding multi-view video using the same, and an encoder and a decoder using the same. In particular, the invention relates to a method of computing a disparity, a method of synthesizing an interpolation view, a method of encoding and decoding multi-view video using the same, and an encoder and a decoder using the same, which can rapidly compute an initial disparity of a block using region segmentation, accurately compute a disparity of the block using a variable block, and synthesize an interpolation view on the basis of a disparity value computed in a pixel basis using an adaptive search range, thereby improving quality of the interpolation view, and also can encode and decode a multi-view video independently from an existing prediction mode while using the interpolation view as a reference picture, thereby improving coding efficiency.

Abstract: A depth image coding method may calculate a depth offset of a depth image, may generate a prediction mode based on the depth offset, may minimize a prediction error of the depth image having a low correlation between adjacent points of view and a low temporal correlation and may enhance a compression rate. The depth offset may be calculated based on a representative value of adjacent pixels included in a template as opposed to using a depth representative value of pixels in a block and header information may not be needed to encode an offset and the offset may be generated by a depth image decoding apparatus. When a plurality of objects is included in a block, a depth offset is calculated for each of the plurality of objects and a motion vector is calculated for each of the plurality of objects and the depth image may be accurately predicted.

Abstract: There are provided a method and an apparatus for generating a multi-viewpoint depth map, and a method for generating a disparity of a multi-viewpoint image. A method for generating a multi-viewpoint depth map according to the present invention includes the steps of: (a) acquiring a multi-viewpoint image constituted by a plurality of images by using a plurality of cameras (b) acquiring an image and depth information by using a depth camera; (c) estimating coordinates of the same point in a space in the plurality of images by using the acquired depth information; (d) determining disparities in the plurality of images with respect to in the same point by searching a predetermined region around the estimated coordinates; and (e) generating a multi-viewpoint depth map by using the determined disparities.

Abstract: The invention provides a method of encoding a multi-view video and an encoder capable of calculating global disparity between pictures at different viewpoints in the multi-view video and forming a hierarchical B picture in both view and time directions on the basis of the calculated global disparity, thereby improving encoding efficiency. According to an embodiment of the invention, there is provided an encoder for encoding data for pictures captured at different viewpoints in a multi-view video. When a picture at one of the first and last viewpoints is encoded into an I picture, the encoder encodes a picture that is captured at the other viewpoint but is disposed in the same time direction into an VP picture.

Abstract: The present invention relates to a method and apparatus for encoding a multi-view video and a method and apparatus for decoding a multi-view video, and more particularly, to a method and apparatus for encoding a multi-view video and a method and apparatus for decoding a multi-view video that can perform encoding and decoding in consideration of global disparity between pictures captured at two different viewpoints to remove spatial redundancy. An embodiment of the invention provides an encoder for compressing data for pictures captured at different viewpoints using global disparity in a multi-view video. The encoder includes: an input unit that receives pictures captured at two or more viewpoints; a reference picture generating unit that generates a new reference picture on the basis of the global disparity between a target picture and a reference picture; and a motion predicting unit that calculates motion vectors of the target picture on the basis of the reference picture.

Abstract: The present invention relates to a device and a method for estimating a depth map, and a method for making an intermediate image and a method for encoding multi-view video using the same.

Abstract: The present invention relates to a method and device for generating a depth image, a method for encoding/decoding the depth image, an encoder/decoder for the same, and a recording medium recording an image generated by the method, which are related to a depth image encoding method that can effectively reduce a bit generation rate using a reference image obtained by at least one camera and improve encoding efficiency. A depth image generating method according to an embodiment of the invention includes a step (a) of obtaining a depth image at a viewpoint and setting the obtained depth image to a reference image; a step (b) of applying a 3D warphing method to the reference image and predicting and generating a depth image at a specific viewpoint; and a step (c) of removing a hole that exists in the predicted and generated depth image.

Abstract: In a method for image prediction of a multi-view video codec, a disparity vector value of a current macroblock is obtained and a corresponding macroblock is searched through the obtained disparity vector value. An intra prediction coding direction value of the searched macroblock is determined as an intra prediction coding direction value of the current macroblock. The current macroblock is coded/decoded through the determined intra prediction coding direction value. The coding efficiency can be increased by reducing a coding amount by omitting a process of obtaining an unnecessary disparity value by using a different-view intra prediction coding direction as a current-view intra prediction coding direction. The coding efficiency can be increased by reducing a coding amount by differentially coding a similar disparity vector. The reliability can be increased through an accurate disparity using multiples of 8 rather than 16, in motion skip for sharing the format of a macroblock.

Abstract: The invention relates to a method of computing a disparity, a method of synthesizing an interpolation view, a method of encoding and decoding multi-view video using the same, and an encoder and a decoder using the same. In particular, the invention relates to a method of computing a disparity, a method of synthesizing an interpolation view, a method of encoding and decoding multi-view video using the same, and an encoder and a decoder using the same, which can rapidly compute an initial disparity of a block using region segmentation, accurately compute a disparity of the block using a variable block, and synthesize an interpolation view on the basis of a disparity value computed in a pixel basis using an adaptive search range, thereby improving quality of the interpolation view, and also can encode and decode a multi-view video independently from an existing prediction mode while using the interpolation view as a reference picture, thereby improving coding efficiency.

Abstract: Provided are a method for image prediction of a multi-view video codec capable of improving coding efficiency, and a computer readable recording medium therefor. The method for image prediction of a multi-view video codec includes partitioning an image to a plurality of base blocks, acquiring information of reference images which are temporally different, acquiring information of reference images which have different views, and predicting a target block based on the acquired information. Accordingly, an image that is most similar to an image of a view to be currently compressed is generated using multiple images of different views, so that coding efficiency can be improved.