Real-time Geometry Corporation

Viewpoint Corporation

Enliven Marketing Technologies Corporation

Identix Corporation

Dg Fastchannel, Inc

Michael Petrov

Alexei Lebedev

Alexander A Migdal

Veronika Shelyekhova

Vadim Abadjev

Leonid Polonskiy

Michael Brauckmann

Vladimir Bernstein

Alexander Zhilyaev

Ernst Mucke

Andrei Afanassenkov

Biometric search methods, apparatus, and systems using operator input for an operator-assisted iterative biometric search are provided. In certain embodiments, a set of candidate records is received at a computing device. At least a first portion of the set of candidate records is displayed. An indication that a first candidate record in the set of candidate records is not a match to the probe record is received. A first search of the set of candidate records using the first candidate record is performed. A score assigned to a second candidate record in the set of candidate records is modified when the second candidate record is more likely to be a match to the first candidate record than a match to the probe record. At least a second portion of the set of candidate records is displayed.

Operator-assisted iterative biometric search

Improvement of the performance, usability, and functionality of software for editing three dimensional computer models and textures is described. Editing operations are performed on the model in an ordered list to reduce the amount of memory required to maintain previous versions, and certain editing operations are precalculated to reduce the amount of real time calculation required. Improvements to usability and functionality are provided by more precisely selecting portions of a model for editing, allowing textures to be moved more easily on the model, and allowing better blending of the appearance of adjacent textures.

Tools for 3D mesh and texture manipulation

System and method for constructing a 3D model of an object based on a series of silhouette and texture map images. In the exemplary embodiment an object is placed on a rotating turntable and a camera, which is stationary, captures images of the object as it rotates on the turntable. In one pass, the system captures a number of photographic images that will be processed into image silhouettes. In a second pass, the system gathers texture data. After a calibration procedure (used to determine the camera's focal length and the turntable's axis of rotation), a silhouette processing module determines a set of two-dimensional polygon shapes (silhouette contour polygons) that describe the contours of the object. The system uses the silhouette contour polygons to create a 3D polygonal mesh model of the object. The system determines the shape of the 3D model analytically by finding the areas of intersection between the edges of the model faces and the edges of the silhouette contour polygons. The system creates an initial, (rough) model of the 3D object from one of the silhouette contour polygons, then executes an overlaying procedure to process each of the remaining silhouette contour polygons. In the overlaying process, the system processes the silhouette contour polygons collected from each silhouette image, projecting each face of the (rough) 3D model onto the image plane of the silhouette contour polygons. The overlaying of each face of the (rough) 3D model onto the 2D plane of the silhouette contour polygons enables the present invention to determine those areas that are extraneous and should be removed from the (rough) 3D model. As the system processes the silhouette contour polygons in each image it removes the extraneous spaces from the initial object model and creates new faces to patch 'holes.' The polygonal mesh model, once completed, can be transformed into a triangulated mesh model. In a subsequent step, the system uses a deterministic procedure to map texture from the texture images onto the triangles of the 3D mesh model, locating that area in the various texture map images that is 'best' for each mesh triangle.

System and method of three-dimensional image capture and modeling

Systems and methods for identifying the contour of an object that is depicted in an image. An object is placed on a rotating turntable and a camera, which is stationary, captures images of the object as it rotates on the turntable. In one pass, the system captures a number of photographic images that will be processed into image silhouettes. After a calibration procedure (used to determine the camera's focal length and the turntable's axis of rotation), a silhouette processing module determines a set of two-dimensional polygon shapes (silhouette contour polygons) that describe the contours of the object by comparing light intensity values in the images of the object with light intensity values from an image of the background without the object.

System and method for constructing a 3D model of an object based on a series of silhouette and texture map images. In the exemplary embodiment an object is placed on a rotating turntable and a camera, which is stationary, captures images of the object as it rotates on the turntable. In one pass, the system captures a number of photographic images that will be processed into image silhouettes. In a second pass, the system gathers texture data. After a calibration procedure (used to determine the camera's focal length and the turntable's axis of rotation), a silhouette processing module determines a set of two-dimensional polygon shapes (silhouette contour polygons) that describe the contours of the object. The system uses the silhouette contour polygons to create a 3D polygonal mesh model of the object. The system determines the shape of the 3D model analytically-by finding the areas of intersection between the edges of the model faces and the edges of the silhouette contour polygons. The system creates an initial, (rough) model of the 3D object from one of the silhouette contour polygons, then executes an overlaying procedure to process each of the remaining silhouette contour polygons. In the overlaying process, the system processes the silhouette contour polygons collected from each silhouette image, projecting each face of the (rough) 3D model onto the image plane of the silhouette contour polygons. The overlaying of each face of the (rough) 3D model onto the 2D plane of the silhouette contour polygons enables the present invention to determine those areas that are extraneous and should be removed from the (rough) 3D model. As the system processes the silhouette contour polygons in each image it removes the extraneous spaces from the initial object model and creates new faces to patch 'holes.' The polygonal mesh model, once completed, can be transformed into a triangulated mesh model. In a subsequent step, the system uses a deterministic procedure to map texture from the texture images onto the triangles of the 3D mesh model, locating that area in the various texture map images that is 'best' for each mesh triangle.

The present invention provides a system for illuminating an object with a special kind of structured light pattern, recording the shape of the reflected points of light by means of a camera, and, by a triangulation technique that does not depend on the fixed direction of the light source relative to the camera, reconstructing the 3D shape of the object through a computer using the data points collected from the reflection of the structured light pattern. The scanning system is portable and does not require data processing contemporaneous with the data collection. The portable system stores in the storage media several images of the objects with different illumination patterns. The data is subsequently processed, by a computer system which applies data processing routines, i.e., the model building algorithms which provide 3D surface generation. The data acquisition according to the present invention is simplified to acquiring of only two or, optionally, four images of the object, thereby significantly increasing the digitization speed over that of laser-based scanners. The light source projects both structured light and uniform illumination light from the same apparent source, and that allows for numerical normalization of the images.

Structured-light, triangulation-based three-dimensional digitizer

System and method for rapid collection of data points and construction of a computer model based on a multi-resolution mesh to describe the surface contours and color of an object. The system collects data by projecting shapes of light against the object and collecting (from a position of triangulation relative to the light projector) images of the light as it reflects from the object. The system and method of the present invention are comprised of a multiple laser stripe generation process to project a number of light shapes against the object, the collection of one or more reflected laser stripes in a single image of an images collector to gather sufficient information to reproduce points on the surface of the object. The system compresses the data associated with collected points, which allows for accurate values for the contours of the object to subpixel accuracy. A multi-resolution analysis, which maintains more points to resolve fine details and removes points to further smooth regions of the objects, leads to significant data compression. The adaptive mesh, consisting of the connection of polygonal finite elements, is automatically generated by the system and is comprised to create multi-resolution meshes at different tolerances. The system and method of the present invention is portable, processes rapidly enough to take 3-D images of animate objects, can be implemented using commercially available equipment, and produces accurate models of objects.

System and method for rapid shape digitizing and adaptive mesh generation

System and method for rapid collection of data points and construction of a computer model based on a multi-resolution mesh to describe the surface contours and color of an object. The system collects data by projecting shapes of light against the object and collecting (from a position of triangulation relative to the light projector) images of the light as it reflects from the object. The system and method of the present invention are comprised of a multiple laser stripe generation process to project a number of light shapes against the object, the collection of one or more reflected laser stripes in a single image of an images collector to gather sufficient information to reproduce points on the surface of the object.

System and method for rapid shaped digitizing and adaptive mesh

A system for high accuracy calibration of a scanning system, which provides an object for calibration of a known geometry, such as a plane or a sphere. Upon that object is marked a pattern of points or geometric shapes, such as a set of equilateral triangles or their vertices. A set of computer programmed elements process photographic images of the object (containing a depiction of the pattern of geometric objects) and scanned images of the object to calibrate to high precision parameters, including a focal point relationship value for scanning systems which use a camera as an image collector (which describes the relationships between the focal point of the lens and the light/image collection device of the camera), a set of values to correct for distortion of the lens, the relative distance between the image collector and the scanner light source and also an initial angle to determine light position scanning. The known geometric constraints of the object allow the location of the object to be determined by photographic images. By scanning the object and computing 3D X,Y,Z coordinates for those points against a constraint that the X,Y,Z coordinates fit onto the surface of the object, the system calibrates parameters pertaining to the relative distances between the image collector and light source, an initial angle of rotation for systems which employ movable light sources and other parameters. The system is applied to multiple camera scanning systems to calibrate rotation matrices and translation vectors which facilitate the joining of the scanned data. The system further facilitates the design of flexible and movable scanning systems, such as portable scanners.

Modular digital audio system having individualized functional modules

A portable 3D scanning system collects 2D-profile data of objects using a combination of a laser-stripe positioning device and a video camera which detects the images of the laser stripe reflected from the object. The scanning system includes a laser-stripe generator, a video camera, a scanning mirror attached to a continuously rotating motor, an encoder or a photodiode operationally coupled to the motor, and associated electronics. As the rotating, scanning mirror reflects the laser stripe and variably positions the laser stripe across the object, the encoder or the photodiode generates signals indicating the angular position of the mirror. The video images of the reflected laser stripes are stored on a storage medium, while data relating to the angular positions of the laser stripes recorded in the video images are simultaneously stored on a storage medium. A computer subsequently synchronizes and processes the recorded laser stripe data with the angular-position data to generate a 3D model of the object by applying triangulation calculation and other post-scanning methods, e.g., multi-resolution analysis and adaptive-mesh generation. The multi-resolution analysis, which applies more points to resolve fine details and fewer points for smooth regions of the objects, leads to significant data compression. The adaptive mesh, which include connected polygonal elements and which may have multiple resolutions and tolerances, is generated by the adaptive-mesh generating routine.

Growing community of inventors

Princeton, NJ, United States of America

Michael Petrov