New York University

Ondotek, Inc.

Siemens Corporate Research, Inc.

Davi Geiger

Hiroshi Ishikawa

Salvatore Paxia

Arash Baratloo

Zvi M Kedem

Peter Wyckoff

Marie-Pierre Jolly

Alok Gupta

Jong Oh

Mark Pipes

A method and system for extracting motion and structure from a sequence of images stored on a computer system, the method comprises obtaining images including a set of three-dimensional (3D) points over a plurality of frames, determining an instantaneous motion of the set of 3D points by an angular velocity and a translation with respect to an axis of rotation, computing an optical flow using the instantaneous motion of the set of 3D points based on a projection of velocity of the set of 3D points, computing a depth of the set of 3D points from the optical flow, and determining an epipolar line in the images using the optical flow.

System and method for determining motion and structure from optical flow

A method is disclosed to automatically segment 3D and higher-dimensional images into two subsets without user intervention, with no topological restriction on the solution, and in such a way that the solution is an optimal in a precisely defined optimization criterion, including an exactly defined degree of smoothness. A minimum-cut algorithm is used on a graph devised so that the optimization criterion translates into the minimization of the graph cut. The minimum cut thus found is interpreted as the segmentation with desired property.

Method and apparatus for segmenting an image in order to locate a part thereof

A process and system are provided for determining a most likely combination of characters. In particular, character data which includes information indicative of at least one handwritten character is obtained. The character data includes at least one set of segmentation points for the handwritten character. Then, a score can be provided for each particular character of a set of previously stored characters based on a comparison between the character data and the previously stored particular character. In addition, it is possible to compare visual aspects of the handwritten character to visual aspects of each of the previously stored characters for determining likely characters. Also, a Fisher Matching procedure can be used on the character data to ascertain the likely characters. A plurality of handwritten characters can include a first character that is connected to a second character of the handwritten characters via a ligature. A further score of the ligature can be determined based on a starting point of the ligature and an ending point of the ligature. Furthermore, the first character can be connected to the second character via a transition. A particular score of the transition can be ascertained based on a difference in length of the first character and the second character. This particular score can be combined with a score corresponding to the particular handwritten characters.

System, process and software arrangement for recognizing handwritten characters

A data image management system (DIMS) that includes a local data image manager (LDIM), a remote data image manager (RDIM), and a remote persistent storage device (RPSD). The LDIM communicates with the RDIM through a direct communication link or through a communication network. The RDIM can store data on and retrieve data from the RPSD. In an environment where an LDIM has been installed in a computer having a &ldquo;local&rdquo; persistent storage device (LPSD), the DIMS allows for the storing of the LPSD's data image on the RPSD, with the LPSD serving as a persistent, consistent cache of the data image. The data image stored on the RPSD is referred to as the &ldquo;master data image&rdquo; and the data image cached on the LPSD is referred to as the &ldquo;local data image&rdquo; or &ldquo;cached data image.&rdquo; The LDIM functions to intercept read/write requests that are intended to be received by the LPSD. The read/write requests specify an address of the LPSD. Upon intercepting a read request, the LDIM is programmed to determine whether the portion of the cached data image that is stored at the specified address is up-to-date. If it is up-to-date, the LDIM retrieves the requested data from the LPSD and passes the data back to the component or device from which it received the request. If it is not up-to-date, the LDIM transmits the read request to the RDIM. Upon receiving the read request, the RDIM locates and reads the requested data from the master data image stored on the RPSD and then transmits the data back to the LDIM.

Data image management via emulation of non-volatile storage device

A technique of tracking an object of interest in a sequence of images using active polyline contours. An image processor converts a sequence of images into digital image data related to light intensity at the pixels of each image. A computer stores the digital image data and forms an initial polyline that substantially outlines an edge of the object in a first image. The computer forms input polylines for each of the subsequent images which substantially outline the edge in the corresponding images and are derived from the optimal polyline of each previous such image. The computer processes the digital image data, performing a graph exploration procedure that starts with the initial polyline in the first image and the input polylines in the subsequent images. The graph exploration procedure searches polylines in a predefined search space to find the corresponding optimal polylines. The computer evaluates edge strength of the different polylines with respect to the light intensity of its underlying pixels to obtain corresponding contour costs. The polyline with the smallest contour cost is selected as the optimal contour for each of the images. The set of optimal contours are used to track the object of interest.

Growing community of inventors

New York, NY, United States of America

Davi Geiger