Microsoft Technology Licensing, LLC

Philip H S Torr

Andrew Blake

Bernhard Schoelkopf

Sami Romdhani

Antonio Criminisi

Jamie Shotton

Richard S Szeliski

Ingemar J Cox

Padmananbhan Anandan

Colin Davidson

A multi-layer graph for dense stereo dynamic programming can improve synthesis of cyclopean virtual images by distinguishing between stereo disparities caused by occlusion and disparities caused by non-fronto-parallel surfaces. This distinction can be leveraged to reduce image artifacts, such as 'halos'. Distinguishing at least between these two types of disparities allows improved matching of left and right pixel data, which increases the amount of correct pixel information used in constructing the cyclopean virtual image and minimizes occlusion artifacts.

Gaze manipulation

A multi-layer graph for dense stereo dynamic programming can improve synthesis of cyclopean virtual images by distinguishing between stereo disparities causes by occlusion and disparities caused by non-fronto-parallel surfaces. In addition, cyclopean virtual image processing may be combined with simulation of three-dimensional translation of a virtual camera to assist in aligning the user's gaze with the virtual camera. Such translation may include without limitation one or more of the following: horizontal (e.g., left and right) translation of the virtual camera, vertical translation (e.g., up and down) of the virtual camera, and axial translation (e.g., toward the subject and away from the subject) of the virtual camera.

Virtual camera translation

A cyclopean virtual imaging system provides a view from a virtual camera placed symmetrically or midway between the calibrated cameras of a physical stereo pair. Such a system may be used in video conferencing applications as well as other stereo image contexts. By combining the left and right stereo images captured by the stereo pair, a virtual image is produced such that the subject appears to be looking into a cyclopean virtual camera, rather than to the left or right of a single physical camera. The cyclopean virtual image is generated by a forward-backward algorithm using a probabilistic distribution of possible disparity families.

Cyclopean virtual imaging via generalized probabilistic smoothing

Systems and methods for object or pattern detection that use a nonlinear support vector (SV) machine are described. In the illustrated and described embodiment, objects or patterns comprising faces are detected. The decision surface is approximated in terms of a reduced set of expansion vectors. In order to determine the presence of a face, the kernelized inner product of the expansion vectors with the input pattern are sequentially evaluated and summed, such that if at any point the pattern can be rejected as not comprising a face, no more expansion vectors are used. The sequential application of the expansion vectors produces a substantial saving in computational time.

Pattern detection using reduced set vectors

Pattern detection

Pattern detection methods and systems and face detection methods and systems

Pattern detection methods and systems, and face detection methods and systems

Feature correspondence between images using an image pyramid is disclosed. In one embodiment, a fundamental matrix between a first and a second image is generated from the image pyramid. The image pyramid is first generated, and has a predetermined number of fineness levels, from a coarsest to a finest level. Each of the images has significant features at each pyramid level. A plurality of hypotheses, or particles, is generated for the fundamental matrix at the coarsest level, based on matching significant features of the images at the coarsest level. In an iterative procession through the levels of the image pyramid, starting at a present level initially set to the coarsest level and then subsequently advanced by one fineness level upon each iteration, an importance sampling function is first formulated from the hypotheses. The plurality of hypotheses is then generated at the next pyramid level based on the function, and on the significant features of the images at this next level. The iteration is complete when the next level has reached the finest pyramid level. The hypotheses generated at the finest level encapsulate the fundamental matrix together with its uncertainty.

Feature correspondence between images using an image pyramid

Automated layer extraction from 2D images making up a 3D scene, and automated image pixel assignment to layers, to provide for scene modeling, is disclosed. In one embodiment, a computer-implemented method determines a number of planes, or layers, and assigns pixels to the planes. The method can determine the number of planes by first determining the high-entropy pixels of the images, and then determining a 1-plane through a predetermined n-plane estimation, such as via a robust estimation, and a most likely x-plane estimation, where x is between 1 and n, such as via a Bayesian approach. Furthermore, the method can assign pixels via an iterative EM approach based on classifying criteria.

Growing community of inventors

Cambridge, United Kingdom

Philip H S Torr