According to one example for segmenting image data, image data comprising color pixel data, IR data, and depth data is received from a sensor. The image data is segmented into a first list of objects based on at least one computed feature of the image data. At least one object type is determined for at least one object in the first list of objects. The segmentation of the first list of objects is refined into a second list of objects based on the at least one object type. In an example, the second list of objects is output.

An image processing apparatus includes a processing circuitry configured: to generate an enhanced image in which an object is enhanced from each of sequentially-generated X-ray images; to detect the position of the object included in each of the generated enhanced images; to determine, while using the position of the object detected from a specific one of the enhanced images as a reference position, processing details of a correcting process performed to arrange the detected position of the object to match the reference position, with respect to each of the enhanced images generated after the specific enhanced image; to sequentially generate corrected images by performing the correcting processes according to the processing details determined for each of the enhanced images, on the X-ray images from which the enhanced images were generated; and to cause a display to display the sequentially-generated corrected images as a moving picture.

An image contrast enhancement method and an apparatus thereof are disclosed, which calculate the degree of influencing the clarity according to the influence feature (e.g., heavy fog, dust, smoke, or etc.) in the image, and then adjust the brightness of the pixels corresponding to features of influencing the clarity according to the degree, thereby enhancing image contrast and removing phenomenon of influencing the clarity in the image.

Disclosed are a method and apparatus for processing a 3D model. To preserve fine structures while de-noising a 3D mesh model, the local structural information around a vertex is captured when designing a de-noising filter. In particular, for a current vertex to be processed, a path, for example, a geodesic path, is determined between the current vertex and each neighboring vertex. For each mesh edge along the path, local variations are calculated for the two end vertices of the mesh edge using a covariance matrix, and a geometric variation for the mesh edge is calculated as the difference between the two local variations. Then structural information for the region between the current vertex and a neighboring vertex is calculated as a function of the geometric variations for mesh edges along the path, for example, as the maximum geometric variation along the path. The present principles can also be adjusted to be used in de-noising 3D point-based models.

Systems, methods and computer storage media for enhancing object representations using data items selected based upon inferred user intents are provided. Information indicative of an initial object representation is received. At least one user intent related to the initial object representation is inferred based upon one or more of a plurality of artifacts related to the user and/or the user's activities. Based upon the inferred user intent and the initial object representation, one or more data items are determined that are relevant to the both the intent and the representation. The determined data items are mapped to the initial object representation to create an enhanced object representation.

First visual information defining the visual content in a first projection may be accessed. Second visual information defining lower versions of the visual content in the first projection may be accessed. A transformation of the visual content from the first projection to a second projection may be determined. The transformation may include a visual compression of a portion of the visual content in the first projection. The portion may be identified. An amount of the visual compression of the portion may be determined. One or more lower resolution versions of the visual content may be selected. The visual content may be transformed using the one or more lower resolution versions of the visual content.

Means which enables image conversion in which a plurality of conversion results can be output without once saving all of video-image data, which has been input from image-pickup means, in a storage medium is provided. A single line memory having a plurality of lines is used while switching the role thereof for a reading use by a video-image converting means and a use for inputting image data from the image-pickup means. The image converting means obtains an input image, which is in the line memory, and carries out conversion of the input image based on a conversion specifying means interpreted by an instruction decoder.

Introduced herein are a variety of techniques for displaying virtual and augmented reality content to a user through head-mounted display (HMD). The techniques described herein can be used to improve the effectiveness of the HMD, as well as general experience and comfort of a user of the HMD. For example, a binocular HMD system could modify the overlap of digital content in real-time so that the two separate images can be viewed without visual discomfort. The HMD system could also present visual stabilizers to each eye allow the user to more easily visually align the two separate images when viewed together. Techniques such as these decrease the eye fatigue and strain experienced by the user when viewing virtual or augmented reality content on HMDs.

Disclosed is an image sensing device including a plurality of defect detectors each suitable for detecting whether a corresponding target image value is defective, and generating detection information corresponding to a result of the detection, a defect scheduler suitable for sequentially outputting one or more defective image values, which are the target image values detected as defective among the plurality of target image values, based on the detection information, and a defect corrector suitable for correcting the output defective image values.

The present disclosure provides a method for enhancing image contrast, including: dividing an input image into a plurality of input sub-images, each of the plurality of input sub-images corresponding to a gray value distribution range, the gray value distribution ranges of the respective input sub-images being different from each other, and each input sub-image comprising all pixels, among pixels of the input image, whose gray values are within the gray value distribution range of the input sub-image; performing an equalization processing on each of the plurality of input sub-images, and obtaining a plurality of output sub-images corresponding to the plurality of input sub-images, respectively, based on a result of the equalization processing; and merging the plurality of output sub-images to obtain an output image corresponding to the input image. The present disclosure further provides a device for enhancing image contrast, a display apparatus, and a computer-readable storage medium.