Abstract: An apparatus for generating a new viewpoint image includes: a new viewpoint image generation unit that shifts the foreground image region in the input image, based on a depth map to generate a new viewpoint image including a missing pixel region; a representative background pixel selection unit that selects background pixels located a predetermined number of pixels away from edge pixels, as representative background pixels; a texture direction map calculation unit that calculates a texture direction of each of the selected representative background pixels, to calculate a texture direction map indicating texture directions, each corresponding to one of missing pixels, based on locations of the representative background pixels and locations of the missing pixels included in the missing pixel region; and a hole-filling processing unit that fills the missing pixel region with the obtained background pixels in accordance with the texture direction map.

Abstract: A human body part detection system includes: a learning mode storing unit storing a learning model; a depth image acquisition unit acquiring a depth image; a foreground human extraction unit extracting a human area; and a human body part detection unit detecting the human body part based on the human area and the learning model. The detection unit calculates a direction of a geodesic path at a first point based on a shortest geodesic path from a base point to a first point, selects a pixel pair at positions obtained after rotating positions of a pixel pair for calculation of the feature in the learning model in accordance with the direction, calculates a feature at the first point based on depth of the selected pair, and determines a label corresponding to the human body part based on the feature at the first point and learning model.

Abstract: An image generation apparatus includes: a pixel-shifted image generation unit which generates a first pixel-shifted image resulting from a process of shifting, based on an input image and a depth map, a pixel in the input image by a shift amount corresponding to the new viewpoint, and a second pixel-shifted image resulting from a process of shifting, by the same shift amount as for the first pixel-shifted image, a pixel in a background edge image included in the input image and containing a pixel located in a background region around an edge of a foreground object; an image fusion unit which fuses the second pixel-shifted image and the first pixel-shifted image to include a background region in the second pixel-shifted image into a hole; and a hole filling unit which fills the hole with pixel information based on the fused pixel-shifted image.

Abstract: A motion estimation apparatus according to an aspect of the present invention is a motion estimation apparatus for estimating, using a set of multi-focus images which correspond to a single scene and have different focuses, motion for each of the first regions included in the scene, including: a cost value computing unit configured to compute, using the set of multi-focus images, for each of the first regions, a cost value indicating a difference between a blur amount of the first region and a standard blur amount determined for each of distances in a depth direction; and a motion estimation unit configured to estimate, using the cost value, the motion of the first region corresponding to the cost value.

Abstract: An image capturing device includes: a pre-capturing module which captures an imaging target at each of different focus positions in a predetermined focus range and outputs, as a capturing result, a plurality of pre-captured images lower in resolution than the output image; an object sharpness evaluating module which computes a sharpness level of each of the pre-captured images captured by the pre-capturing module; a focus varying range determining module which determines a focus position varying range within the predetermined focus range based on sharpness levels computed by the object sharpness evaluating module, such that a signal-to-noise ratio of the output image is greater than or equal to a predetermined threshold; and an image capturing module which captures the imaging target while varying the focus position according to the focus position varying range determined by the focus varying range determining module, and outputs the output image as a capturing result.

Abstract: A depth information generating device includes a region extracting unit that detects a human face in at least one two-dimensional image, and based on the detected face, extracts a human figure region indicating a human figure within a region of the at least one two-dimensional image; and a depth map generating unit that gives a depth value different from a depth value of a region other than the human figure region to the human figure region to generate a depth map that separates the human figure region from the region other than the human figure region.

Abstract: A stereoscopic image aligning apparatus (200) automatically aligns image pairs for stereoscopic viewing in a shorter amount of time than conventional apparatuses, which is applicable to image pairs captured by a single sensor camera or a variable baseline camera, without relying on camera parameters.

Abstract: The current invention provides methods for 3D content capturing, 3D content coding and packaging at content production side, and 3D content consuming and rendering at display or terminal side, in order to ensure healthy and effective 3D viewing all the time. According to the current invention Maximum disparity and 3D budget which are scene dependent, are calculated, utilized for coding and embedded in the coded streams or media file, and checked against the allowable values during the content rendering, so to determine if the same 3D content can be shown to the user according to the viewing condition which the user has. In case it cannot meet healthy 3D viewing guideline, it is suggested to adjust the 3D content for its new maximum disparity and 3D budget to be within the allowable range, to achieve healthy and effective 3D viewing for that user with his/her viewing condition.

Abstract: A distance estimation device (1x) prevents the variation in distance estimation accuracy that has been a problem in distance estimation performed using an all-in-focus image and an arbitrary-focal-depth image. The distance estimation device can estimate a focal distance of any subject with a high degree of accuracy, and includes: a first generation unit (10w) generating a first distance map indicating a focal depth determined from an all-in-focus image and a first focal-depth image; a second generation unit (11w) generating a second distance map indicating a focal depth determined from the all-in-focus image and a second focal-depth image; and a distance map synthesis unit (14) synthesizing the generated first distance map and the generated second distance map to generate a synthetic distance map.

Abstract: A method and apparatus for motion blur and ghosting prevention in imaging system is presented. A residue image is computed by performing spatial-temporal filter with a set of absolute image difference of image pairs from input images. A noise adaptive pixel threshold is computed for every pixel based on noise statistics of image sensor. The residue image and the noise adaptive pixel threshold are used to create a motion masking map. The motion masking map is used to represent motion and non-motion pixels in pixels merging. The pixels merging step is performed to generate an output image by considering the motion pixels where the motion pixels are performed separately. The resulting output image having no or less motion blur and ghosting artifacts can be obtained, even the input images having different degree of motion blur between each of the image, while the complexity is low. It is preferred that the current invention is applied in the Bayer raw domain.

Abstract: An image generation apparatus includes: a pixel-shifted image generation unit which generates a first pixel-shifted image resulting from a process of shifting, based on an input image and a depth map, a pixel in the input image by a shift amount corresponding to the new viewpoint, and a second pixel-shifted image resulting from a process of shifting, by the same shift amount as for the first pixel-shifted image, a pixel in a background edge image included in the input image and containing a pixel located in a background region around an edge of a foreground object; an image fusion unit which fuses the second pixel-shifted image and the first pixel-shifted image to include a background region in the second pixel-shifted image into a hole; and a hole filling unit which fills the hole with pixel information based on the fused pixel-shifted image.

Abstract: A motion estimation apparatus according to an aspect of the present invention is a motion estimation apparatus for estimating, using a set of multi-focus images which correspond to a single scene and have different focuses, motion for each of the first regions included in the scene, including: a cost value computing unit configured to compute, using the set of multi-focus images, for each of the first regions, a cost value indicating a difference between a blur amount of the first region and a standard blur amount determined for each of distances in a depth direction; and a motion estimation unit configured to estimate, using the cost value, the motion of the first region corresponding to the cost value.

Abstract: An image capturing device includes: a pre-capturing module which captures an imaging target at each of different focus positions in a predetermined focus range and outputs, as a capturing result, a plurality of pre-captured images lower in resolution than the output image; an object sharpness evaluating module which computes a sharpness level of each of the pre-captured images captured by the pre-capturing module; a focus varying range determining module which determines a focus position varying range within the predetermined focus range based on sharpness levels computed by the object sharpness evaluating module, such that a signal-to-noise ratio of the output image is greater than or equal to a predetermined threshold; and an image capturing module which captures the imaging target while varying the focus position according to the focus position varying range determined by the focus varying range determining module, and outputs the output image as a capturing result.

Abstract: A depth information generating device according to the present invention includes a region extracting unit that detects a human face in at least one two-dimensional image, and based on the detected face, extracts a human figure region indicating a human figure within a region of the at least one two-dimensional image; and a depth map generating unit that gives a depth value different from a depth value of a region other than the human figure region to the human figure region to generate a depth map that separates the human figure region from the region other than the human figure region. Thereby, a depth map for displaying a 3D pop-out from which a viewer feels no unnaturalness can be generated.

Abstract: A stereoscopic image aligning apparatus (200) automatically aligns image pairs for stereoscopic viewing in a shorter amount of time than conventional apparatuses, which is applicable to image pairs captured by a single sensor camera or a variable baseline camera, without relying on camera parameters.

Abstract: A distance estimation device (1x) prevents the variation in distance estimation accuracy that has been a problem in distance estimation performed using an all-in-focus image and an arbitrary-focal-depth image. The distance estimation device can estimate a focal distance of any subject with a high degree of accuracy, and includes: a first generation unit (10w) generating a first distance map indicating a focal depth determined from an all-in-focus image and a first focal-depth image; a second generation unit (11w) generating a second distance map indicating a focal depth determined from the all-in-focus image and a second focal-depth image; and a distance map synthesis unit (14) synthesizing the generated first distance map and the generated second distance map to generate a synthetic distance map.

Abstract: A method and apparatus for motion blur and ghosting prevention in imaging system is presented. A residue image is computed by performing spatial-temporal filter with a set of absolute image difference of image pairs from input images. A noise adaptive pixel threshold is computed for every pixel based on noise statistics of image sensor. The residue image and the noise adaptive pixel threshold are used to create a motion masking map. The motion masking map is used to represent motion and non-motion pixels in pixels merging. The pixels merging step is performed to generate an output image by considering the motion pixels where the motion pixels are performed separately. The resulting output image having no or less motion blur and ghosting artifacts can be obtained, even the input images having different degree of motion blur between each of the image, while the complexity is low. It is preferred that the current invention is applied in the Bayer raw domain.