Patents by Inventor Sami Romdhani
Sami Romdhani has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11132531Abstract: The invention relates to a method of face detection and determination of face pose from a three-dimensional mesh comprising the steps of: determining (200) a curvature map of the three-dimensional surface, according to an average curvature or a Gaussian curvature, detecting (300), within the curvature map, a facial nose tip (B) where the three-dimensional facial surface has a predetermined curvature, determining (400) a face pose defined by coordinates of the nose tip (B) and by three angles of rotation of the face around three axes, delineating a face zone and comparison (450) between the delineated face zone and a reference face template, in order to validate the detection of a real face in the three-dimensional mesh. The invention also relates to a method for checking the identity of an individual, using a frontalized shot of their face obtained via the method for face detection and determination of pose applied to a not necessarily frontalized shot.Type: GrantFiled: August 23, 2019Date of Patent: September 28, 2021Assignee: IDEMIA IDENTITY & SECURITY FRANCEInventor: Sami Romdhani
-
Publication number: 20200065564Abstract: The invention relates to a method of face detection and determination of face pose from a three-dimensional mesh comprising the steps of: determining (200) a curvature map of the three-dimensional surface, according to an average curvature or a Gaussian curvature, detecting (300), within the curvature map, a facial nose tip (B) where the three-dimensional facial surface has a predetermined curvature, determining (400) a face pose defined by coordinates of the nose tip (B) and by three angles of rotation of the face around three axes, delineating a face zone and comparison (450) between the delineated face zone and a reference face template, in order to validate the detection of a real face in the three-dimensional mesh. The invention also relates to a method for checking the identity of an individual, using a frontalized shot of their face obtained via the method for face detection and determination of pose applied to a not necessarily frontalized shot.Type: ApplicationFiled: August 23, 2019Publication date: February 27, 2020Inventor: Sami ROMDHANI
-
Patent number: 10235814Abstract: The invention relates to a method for generating a three-dimensional facial model the shape of which can changed on the basis of a plurality of images of faces of persons, including the steps that involve: generating a facial template; acquiring shapes from examples of faces of persons; repeatedly changing the shape of the template for each example of a face of a person, so that the shape of the changed template corresponds to the shape of the face example, and determining the change in shape between the initial template and the changed template; and generating the facial model as a linear combination of the shape of the template and the changes in shape between the initial template and the changed template, for each example of a face of a person. The invention also relates to a method for processing an image of a face of a person such as to generate a three-dimensional image of the face of the person from of said deformable model.Type: GrantFiled: November 20, 2013Date of Patent: March 19, 2019Assignee: IDEMIA IDENTITY & SECURITYInventor: Sami Romdhani
-
Publication number: 20150310673Abstract: The invention relates to a method for generating a three-dimensional facial model the shape of which can changed on the basis of a plurality of images of faces of persons, including the steps that involve: generating a facial template; acquiring shapes from examples of faces of persons; repeatedly changing the shape of the template for each example of a face of a person, so that the shape of the changed template corresponds to the shape of the thee example, and determining the change in shape between the initial template and the changed template; and generating the facial model as a linear combination of the shape of the template and the changes in shape between the initial template and the changed template, for each example of a face of a person. The invention also relates to a method for processing an image of a face of a person such as to generate a three-dimensional image of the face of the person from of said deformable model.Type: ApplicationFiled: November 20, 2013Publication date: October 29, 2015Inventor: Sami Romdhani
-
Patent number: 7756325Abstract: Disclosed is an improved algorithm for estimating the 3D shape of a 3-dimensional object, such as a human face, based on information retrieved from a single photograph by recovering parameters of a 3-dimensional model and methods and systems using the same. Beside the pixel intensity, the invention uses various image features in a multi-features fitting algorithm (MFF) that has a wider radius of convergence and a higher level of precision and provides thereby better results.Type: GrantFiled: June 20, 2006Date of Patent: July 13, 2010Assignee: University of BaselInventors: Thomas Vetter, Sami Romdhani, Jean-Sebastian Pierrard
-
Patent number: 7391908Abstract: 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.Type: GrantFiled: February 28, 2005Date of Patent: June 24, 2008Assignee: Microsoft CorporationInventors: Andrew Blake, Sami Romdhani, Bernhard Schoelkopf, Philip H. S. Torr
-
Patent number: 7236626Abstract: 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.Type: GrantFiled: February 28, 2005Date of Patent: June 26, 2007Assignee: Microsoft CorporationInventors: Andrew Blake, Sami Romdhani, Bernhard Schoelkopf, Philip H. S. Torr
-
Publication number: 20070031028Abstract: Disclosed is an improved algorithm for estimating the 3D shape of a 3-dimensional object, such as a human face, based on information retrieved from a single photograph by recovering parameters of a 3-dimensional model and methods and systems using the same. Beside the pixel intensity, the invention uses various image features in a multi-features fitting algorithm (MFF) that has a wider radius of convergence and a higher level of precision and provides thereby better results.Type: ApplicationFiled: June 20, 2006Publication date: February 8, 2007Inventors: Thomas Vetter, Sami Romdhani, Jean-Sebastian Pierrard
-
Patent number: 7099504Abstract: 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.Type: GrantFiled: May 21, 2004Date of Patent: August 29, 2006Assignee: Microsoft CorporationInventors: Andrew Blake, Sami Romdhani, Bernhard Schoelkopf, Philip H. S. Torr
-
Publication number: 20050196048Abstract: 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.Type: ApplicationFiled: February 28, 2005Publication date: September 8, 2005Applicant: Microsoft CorporationInventors: Andrew Blake, Sami Romdhani, Bernhard Schoelkopf, Philip Torr
-
Publication number: 20050157933Abstract: 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.Type: ApplicationFiled: February 28, 2005Publication date: July 21, 2005Applicant: Microsoft CorporationInventors: Andrew Blake, Sami Romdhani, Bernhard Schoelkopf, Philip Torr
-
Publication number: 20040213439Abstract: 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.Type: ApplicationFiled: May 21, 2004Publication date: October 28, 2004Applicant: Microsoft CorporationInventors: Andrew Blake, Sami Romdhani, Bernhard Schoelkopf, Philip H.S. Torr
-
Patent number: 6804391Abstract: 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.Type: GrantFiled: November 22, 2000Date of Patent: October 12, 2004Assignee: Microsoft CorporationInventors: Andrew Blake, Sami Romdhani, Bernhard Schoelkopf, Philip H. S. Torr