A computer-implemented method for generating a feature descriptor for a location in an image for use in performing descriptor matching in analysing the image, the method comprising determining a set of samples characterising a location in an image by sampling scale-space data representative of the image, the scale-space data comprising data representative of the image at a plurality of length scales; and generating a feature descriptor in dependence on the determined set of samples.

The present invention relates to an obstacle detecting apparatus and an obstacle detecting method and an exemplary embodiment of the present invention provides an obstacle detecting apparatus, including: a stereo camera which photographs a front of a vehicle to generate a left image and a right image; an image matching unit which applies a block matching algorithm to the left image and the right image to extract a plurality of feature points having a similarity which is equal to or larger than a reference value and calculate a coordinate and a depth of the plurality of extracted feature points; a plane calculating unit which connects every three feature points of the plurality of feature points to calculate a plurality of triangles, based on the coordinate and the depth of the plurality of feature points; a normal calculating unit which calculates directions of normal of the plurality of triangles, based on the coordinate and the depth of three feature points corresponding to vertexes of the plurality of triangles; and an obstacle detecting unit which detects an obstacle in front of the vehicle based on a result of comparing a photographing direction of the stereo camera with the normal directions of the plurality of triangles.

Example embodiments may allow for the efficient, edge-preserving filtering, upsampling, or other processing of image data with respect to a reference image. A cost-minimization problem to generate an output image from the input array is mapped onto regularly-spaced vertices in a multidimensional vertex space. This mapping is based on an association between pixels of the reference image and the vertices, and between elements of the input array and the pixels of the reference image. The problem is them solved to determine vertex disparity values for each of the vertices. Pixels of the output image can be determined based on determined vertex disparity values for respective one or more vertices associated with each of the pixels. This fast, efficient image processing method can be used to enable edge-preserving image upsampling, image colorization, semantic segmentation of image contents, image filtering or de-noising, or other applications.

Devices, such as computer readable media, and methods, such as automated methods, for labeling and/or matching. Some of the devices and methods are particularly useful for anatomical labeling of human airway trees. Some of the devices and methods are particularly useful for matching branch-points of human airway trees from represented in two or more graphs.

A method, computer readable storage medium, and system are disclosed for improving communication productivity, comprising: capturing at least one three-dimensional (3D) stream of data on two or more subjects; extracting a time-series of skeletal data from the at least one 3D stream of data on the two or more subjects; and determining an engagement index between the two or more subjects by comparing the time-series of skeletal data on each of the two or more subjects over a time window.

Image processing apparatus programmed to: continuously shoot a subject to obtain images, and detect the object and extract a position of the object from a three-dimensional position of the subject in the images; detect the person and extract a position of the person from the three-dimensional position, and extract, from the position of the person, part information pieces including respective positions of characteristic parts of the person; generate a pose class for each set of part information pieces, the part information pieces being similar to one another in correlation between parts of the person calculated from each of the part information pieces; identify a pose class to which the correlation between parts of the person belongs, among generated pose classes, when a distance between the person and the object is within a predetermined range; and store the identified pose class in association with the object.

Embodiments of the present invention provide a method for obtaining vital sign data of a target object, including: obtaining a 3D depth image of a target object; obtaining, according to depth values of pixels in the 3D depth image of the target object, framework parameters of the target object and a graphic contour of the target object, where the depth value, is obtained according to the distance information, indicates a distance between a point on the target object and the imaging device; retrieving a 3D model matching the framework parameters of the target object and the graphic contour of the target object from a 3D model library, and obtaining a parameter ratio of the 3D model; obtaining at least one real size of the target object; and obtaining vital sign data of the target object according to the parameter ratio of the 3D model and the at least one real size.

Disclosed are a quantum system-based image pattern recognition computation apparatus and method for machine vision and a quantum system-based machine vision apparatus. The computation apparatus recognizes patterns between images in machine vision by using a quantum system. The computation apparatus includes a modeling unit and an interpretation unit. The modeling unit sets up an objective function based on the similarity between a first pattern derived from the relationships between points of interests of a first image and a second pattern derived from the relationships between points of interests of a second image. The interpretation unit finds an optimum first pattern and an optimum second pattern, in which the similarity between the first pattern and the second pattern is optimized, by interpreting a final quantum state obtained through an adiabatic evolution process of the quantum system in which the objective function is optimized.

Disclosed are a method of generating a hierarchical structured pattern based descriptor and a method and a device for recognizing an object in an image using the same. The method of generating a hierarchical structured pattern based descriptor may include generating a hierarchical structured pattern by defining a parent node based on a patch region for a feature point of an input image to be analyzed and defining a child node obtained by dividing the parent node to a predetermined depth, calculating a master direction vector of the patch region based on position coordinates and representative pixel values of the parent node and the child node, and calculating a rotation angle of the patch region based on the master direction vector and rotating the hierarchical structured pattern by the rotation angle.

According to an embodiment, a density measuring device includes a first calculator, a second calculator, and a first generator. The first calculator calculates, from an image including objects of a plurality of classes classified according to a predetermined rule, for each of a plurality of regions formed by dividing the image, density of the objects captured in the region. The second calculator calculates, from the density of the objects captured in each of the regions, likelihood of each object class captured in each of the regions. The first generator generates density data, in which position corresponding to each of the regions in the image is assigned with the density of the object class having at least the higher likelihood than the lowest likelihood from among likelihoods calculated for object classes captured in the corresponding region.