Patents by Inventor COSMIN STAN
COSMIN STAN 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: 11895427Abstract: A method for generating a composite image comprises: detecting a color temperature of a background image; acquiring from a camera through an image signal processor, ISP, performing white balance correction of acquired image data, an image including a foreground region including face of a user; and detecting a color temperature of the foreground region. Responsive to the color temperature for the foreground region differing from that of the background image by more than a threshold amount, a color temperature for white balance correction of a subsequently acquired image is set which causes skin pixels within the foreground region of the subsequently acquired image to have a color temperature closer to the color temperature for the background image. Pixel values of the foreground region are combined with pixel values of the background image corresponding to a background region of the acquired image to provide the composite image.Type: GrantFiled: August 25, 2021Date of Patent: February 6, 2024Inventors: Razvan Condorovici, Andra Stan, Cosmin Stan
-
Publication number: 20240012387Abstract: Systems and methods for presenting video of execution of a robotic process automation (RPA) workflow at a remote computing system are provided. Execution of the RPA workflow by a remote computing system is initiated. Video of the execution of the RPA workflow by the remote computing system is received at a local computing system. The video is presented at the local computing system.Type: ApplicationFiled: July 8, 2022Publication date: January 11, 2024Applicant: UiPath, Inc.Inventors: Gheorghe Cosmin STAN, Mircea GRIGORE, Vasile BUJAC, Arabela-Elena PASLARU, George-Cosmin VLAD
-
Publication number: 20230066958Abstract: A method for generating a composite image comprises: detecting a color temperature of a background image; acquiring from a camera through an image signal processor, ISP, performing white balance correction of acquired image data, an image including a foreground region including face of a user; and detecting a color temperature of the foreground region. Responsive to the color temperature for the foreground region differing from that of the background image by more than a threshold amount, a color temperature for white balance correction of a subsequently acquired image is set which causes skin pixels within the foreground region of the subsequently acquired image to have a color temperature closer to the color temperature for the background image. Pixel values of the foreground region are combined with pixel values of the background image corresponding to a background region of the acquired image to provide the composite image.Type: ApplicationFiled: August 25, 2021Publication date: March 2, 2023Applicant: FotoNation LimitedInventors: Razvan CONDOROVICI, Andra STAN, Cosmin STAN
-
Patent number: 10834331Abstract: A method is disclosed for processing at least a portion of an input digital image comprising rows of pixels extending in two mutually perpendicular directions over a 2D field. The method comprises defining a kernel for processing an image, the kernel comprising at least one row of contiguous elements of the same non-zero value (such rows being referred to herein as equal-valued kernel regions), the equal-valued kernel regions, if more than one, extending parallel to one another. For each pixel in at least selected parallel rows of pixels within the image portion, the cumulative sum of the pixel is calculated by adding a value of the pixel to the sum of all preceding pixel values in the same row of the image portion. The kernel is convolved with the image portion at successive kernel positions relative to the image portion such that each pixel in each selected row is a target pixel for a respective kernel position.Type: GrantFiled: July 3, 2019Date of Patent: November 10, 2020Assignee: FotoNation LimitedInventors: Cosmin Pit Rada, Cosmin Stan
-
Patent number: 10769435Abstract: A method of image processing within an image acquisition device. In one embodiment an image including one or more face regions is acquired and one or more iris regions are identified within the one or more face regions. The one or more iris regions are analyzed to identify any iris region containing an iris pattern that poses a risk of biometrically identifying a subject within the image. Responsive to identifying any such iris region, a respective substitute iris region, containing an iris pattern distinct from the identified iris pattern to avoid identifying the subject within the image, is determined and the identified iris region is replaced with the substitute iris region in the original image.Type: GrantFiled: October 17, 2019Date of Patent: September 8, 2020Assignee: FotoNation LimitedInventors: Ilariu Raducan, Ruxandra Vranceanu, Razvan Condorovici, Cosmin Stan, Peter Corcoran
-
Publication number: 20200050849Abstract: A method of image processing within an image acquisition device. In one embodiment an image including one or more face regions is acquired and one or more iris regions are identified within the one or more face regions. The one or more iris regions are analyzed to identify any iris region containing an iris pattern that poses a risk of biometrically identifying a subject within the image. Responsive to identifying any such iris region, a respective substitute iris region, containing an iris pattern distinct from the identified iris pattern to avoid identifying the subject within the image, is determined and the identified iris region is replaced with the substitute iris region in the original image.Type: ApplicationFiled: October 17, 2019Publication date: February 13, 2020Applicant: FotoNation LimitedInventors: Ilariu RADUCAN, Ruxandra VRANCEANU, Razvan CONDOROVICI, Cosmin STAN, Peter CORCORAN
-
Publication number: 20200007791Abstract: A method is disclosed for processing at least a portion of an input digital image comprising rows of pixels extending in two mutually perpendicular directions over a 2D field. The method comprises defining a kernel for processing an image, the kernel comprising at least one row of contiguous elements of the same non-zero value (such rows being referred to herein as equal-valued kernel regions), the equal-valued kernel regions, if more than one, extending parallel to one another. For each pixel in at least selected parallel rows of pixels within the image portion, the cumulative sum of the pixel is calculated by adding a value of the pixel to the sum of all preceding pixel values in the same row of the image portion. The kernel is convolved with the image portion at successive kernel positions relative to the image portion such that each pixel in each selected row is a target pixel for a respective kernel position.Type: ApplicationFiled: July 3, 2019Publication date: January 2, 2020Applicant: FotoNation LimitedInventors: Cosmin PIT RADA, Cosmin STAN
-
Patent number: 10452910Abstract: A method of image processing within an image acquisition device. In one embodiment an image including one or more face regions is acquired and one or more iris regions are identified within the one or more face regions. The one or more iris regions are analyzed to identify any iris region containing an iris pattern that poses a risk of biometrically identifying a subject within the image. Responsive to identifying any such iris region, a respective substitute iris region, containing an iris pattern distinct from the identified iris pattern to avoid identifying the subject within the image, is determined and the identified iris region is replaced with the substitute iris region in the original image.Type: GrantFiled: March 27, 2015Date of Patent: October 22, 2019Assignee: FotoNation LimitedInventors: Ilariu Raducan, Ruxandra Vranceanu, Razvan Condorovici, Cosmin Stan, Peter Corcoran
-
Patent number: 10348979Abstract: A method is disclosed for processing at least a portion of an input digital image comprising rows of pixels extending in two mutually perpendicular directions over a 2D field. The method comprises defining a kernel for processing an image, the kernel comprising at least one row of contiguous elements of the same non-zero value (such rows being referred to herein as equal-valued kernel regions), the equal-valued kernel regions, if more than one, extending parallel to one another. For each pixel in at least selected parallel rows of pixels within the image portion, the cumulative sum of the pixel is calculated by adding a value of the pixel to the sum of all preceding pixel values in the same row of the image portion. The kernel is convolved with the image portion at successive kernel positions relative to the image portion such that each pixel in each selected row is a target pixel for a respective kernel position.Type: GrantFiled: April 21, 2016Date of Patent: July 9, 2019Assignee: FotoNation LimitedInventors: Cosmin Pit Rada, Cosmin Stan
-
Patent number: 10298949Abstract: A method of processing a stream of images comprises obtaining an image of a scene with a relatively short exposure time (SET) and obtaining an image of the same scene with a relatively longer exposure time (LET). Motion blur characteristics for the SET image corresponding to motion within the LET image are determined and the motion blur characteristics are applied to the SET image. The blurred SET image and the LET image are blended to provide a HDR image. The process is repeated for successive pairs of images in the stream to provide a sequence of HDR images which can be encoded in a video stream.Type: GrantFiled: April 28, 2016Date of Patent: May 21, 2019Assignee: FotoNation LimitedInventors: Razvan G. Condorovici, Cosmin Stan, Ilinca Dobre
-
Publication number: 20170186138Abstract: A method of image processing within an image acquisition device comprises acquiring an image including one or more face regions and identifying one or more iris regions within the one or more face regions. The one or more iris regions are analyzed to identify any iris region comprising an iris pattern of sufficient quality to pose a risk of biometrically identifying a subject within the image. Responsive to identifying any such iris region, a respective substitute iris region comprising an iris pattern sufficiently distinct from the identified iris pattern to avoid identifying the subject within the image is determined and the identified iris region is replaced with the substitute iris region in the original image.Type: ApplicationFiled: March 27, 2015Publication date: June 29, 2017Inventors: Ilariu RADUCAN, Ruxandra VRANCEANU, Razvan CONDOROVICI, Cosmin STAN, Peter CORCORAN
-
Publication number: 20160323596Abstract: A method of processing a stream of images comprises obtaining an image of a scene with a relatively short exposure time (SET) and obtaining an image of the same scene with a relatively longer exposure time (LET). Motion blur characteristics for the SET image corresponding to motion within the LET image are determined and the motion blur characteristics are applied to the SET image. The blurred SET image and the LET image are blended to provide a HDR image. The process is repeated for successive pairs of images in the stream to provide a sequence of HDR images which can be encoded in a video stream.Type: ApplicationFiled: April 28, 2016Publication date: November 3, 2016Inventors: Razvan G. CONDOROVICI, Cosmin STAN, Ilinca DOBRE
-
Publication number: 20160316152Abstract: A method is disclosed for processing at least a portion of an input digital image comprising rows of pixels extending in two mutually perpendicular directions over a 2D field. The method comprises defining a kernel for processing an image, the kernel comprising at least one row of contiguous elements of the same non-zero value (such rows being referred to herein as equal-valued kernel regions), the equal-valued kernel regions, if more than one, extending parallel to one another. For each pixel in at least selected parallel rows of pixels within the image portion, the cumulative sum of the pixel is calculated by adding a value of the pixel to the sum of all preceding pixel values in the same row of the image portion. The kernel is convolved with the image portion at successive kernel positions relative to the image portion such that each pixel in each selected row is a target pixel for a respective kernel position.Type: ApplicationFiled: April 21, 2016Publication date: October 27, 2016Inventors: Cosmin PIT RADA, Cosmin STAN
-
Patent number: 9239957Abstract: A method of image processing within an image acquisition device comprises: acquiring an image including one or more face regions and identifying one or more eye-iris regions within the one or more face regions. The one or more eye-iris regions are analyzed to identify any eye-iris region comprising an eye-iris pattern of sufficient quality to pose a risk of biometrically identifying a person within the image. Responsive to identifying any such eye-iris region, a respective substitute eye-iris region comprising an eye-iris pattern sufficiently distinct from the identified eye-iris pattern to avoid identifying the person within the image is determined, and the identified eye-iris region is replaced with the substitute eye-iris region in the original image.Type: GrantFiled: October 3, 2014Date of Patent: January 19, 2016Assignee: FotoNation LimitedInventors: Ilariu Raducan, Ruxandra Vranceanu, Razvan Condorovici, Cosmin Stan, Peter Corcoran
-
Publication number: 20150286854Abstract: A method of image processing within an image acquisition device comprises: acquiring an image including one or more face regions and identifying one or more eye-iris regions within the one or more face regions. The one or more eye-iris regions are analyzed to identify any eye-iris region comprising an eye-iris pattern of sufficient quality to pose a risk of biometrically identifying a person within the image. Responsive to identifying any such eye-iris region, a respective substitute eye-iris region comprising an eye-iris pattern sufficiently distinct from the identified eye-iris pattern to avoid identifying the person within the image is determined, and the identified eye-iris region is replaced with the substitute eye-iris region in the original image.Type: ApplicationFiled: October 3, 2014Publication date: October 8, 2015Inventors: ILARIU RADUCAN, RUXANDRA VRANCEANU, RAZVAN CONDOROVICI, COSMIN STAN, PETER CORCORAN