Patents by Inventor Soumenkumar Dey
Soumenkumar Dey 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).
-
Publication number: 20140328402Abstract: A method includes determining, through a decoder engine executing on a processor communicatively coupled to a memory and/or a hardware decoder, that a reference video frame of a bi-predicted frame or a predicted frame currently being decoded is unavailable or corrupt. The method also includes initializing, through the decoder engine and/or the hardware decoder, a previously initialized another reference video frame of the bi-predicted frame or the predicted frame as the reference video frame instead of the unavailable or the corrupt reference video frame to enable prediction of the bi-predicted frame or the predicted frame with reduced distortion.Type: ApplicationFiled: May 6, 2013Publication date: November 6, 2014Applicant: NVIDIA CorporationInventors: Shashank Garg, Soumenkumar Dey
-
Patent number: 8787445Abstract: A technique of encoding video frames allocates an available number of bits to different portions of the video frame. A processing unit identifies a region of interest (ROI) in a video frame, and computes a first and second complexity parameter respectively representing the change in video information in the ROI portions and non-ROI portions in the video frame relative to a reference frame. Bits are allocated to the ROI portion proportional (positive correlation) to the first complexity parameter and a ratio of the area of the ROI to the area of the frame. The remaining available bits are allocated to the non-ROI. In an embodiment, the bits are encoded according to H.264 standard.Type: GrantFiled: March 15, 2007Date of Patent: July 22, 2014Assignee: Nvidia CorporationInventors: Vinayak Pore, Soumenkumar Dey, Vikas Patel
-
Patent number: 8442111Abstract: An encoder provided according to an aspect of the present invention uses different encoding techniques depending on an amount of power available in the corresponding durations. Due to the ability to use such different encoding techniques, power may be optimally utilized. The optimization is further enhanced by dynamically switching between encoding techniques according to power amount availability in corresponding durations. In an embodiment, each encoding technique estimates motion vectors at corresponding level of precision (thereby consuming a corresponding level of power) and the precision level is chosen to correspond to available power budget. The circuitry not required for a desired precision level may be switched off.Type: GrantFiled: November 24, 2008Date of Patent: May 14, 2013Assignee: NVIDIA CorporationInventors: Shashank Garg, Vinayak Jayaram Pore, Soumenkumar Dey, Manish Jatashanker Pandey, Harikrishna Madadi Reddy, Manindra Nath Parhy
-
Patent number: 8351776Abstract: Multiple sets of pixel values representing a captured image of a scene are received, with each set representing an image captured with a corresponding degree of focus. An image processor may identify a region of interest in the captured image, automatically determine the configuration parameters for a lens assembly to provide a desired degree of focus for the region of interest, and generate signals to configure a lens assembly. In an embodiment, the region of interest is a face, the desired degree of focus of the face is determined by computing a rate of variation of luminance of pixels representing the face, and the desired degree is the degree of the image having the maximum degree of focus.Type: GrantFiled: December 30, 2009Date of Patent: January 8, 2013Assignee: Nvidia CorporationInventors: Shashank Garg, Vinayak Pore, Soumenkumar Dey
-
Patent number: 8340512Abstract: Multiple sets of pixel values representing a captured image of a scene are received, with each set representing an image captured with a corresponding degree of focus. An image processor may identify a region of interest in the captured image, automatically determine the configuration parameters for a lens assembly to provide a desired degree of focus for the region of interest, and generate signals to configure a lens assembly. In an embodiment, the region of interest is a face, the desired degree of focus of the face is determined by computing a rate of variation of luminance of pixels representing the face, and the desired degree is the degree of the image having the maximum degree of focus.Type: GrantFiled: March 15, 2007Date of Patent: December 25, 2012Assignee: Nvidia CorporationInventors: Shashank Garg, I, Vinayak Pore, II, Soumenkumar Dey, III
-
Patent number: 8290357Abstract: An image processor, which determines appropriate exposure parameters for a shutter assembly in a camera. The image processor may computationally determine a region of interest in a scene sought to be captured, and set the parameters to ensure that the exposure parameters are set to capture an image of the scene with the region of interest having a desired brightness level. In an embodiment, pixel values of multiple frames (each frame with a corresponding set of configuration parameters of the shutter assembly) may be examined to determine the frame having pixel values with the region having the desired brightness level. The shutter assembly may be configured with the parameters corresponding to such a frame to provide an auto-exposure feature.Type: GrantFiled: March 15, 2007Date of Patent: October 16, 2012Assignee: Nvidia CorporationInventors: Soumenkumar Dey, Vinayak Pore, II, Jaiprakash Rao, Ankur Gupta
-
Publication number: 20100128777Abstract: An encoder provided according to an aspect of the present invention uses different encoding techniques depending on an amount of power available in the corresponding durations. Due to the ability to use such different encoding techniques, power may be optimally utilized. The optimization is further enhanced by dynamically switching between encoding techniques according to power amount availability in corresponding durations. In an embodiment, each encoding technique estimates motion vectors at corresponding level of precision (thereby consuming a corresponding level of power) and the precision level is chosen to correspond to available power budget. The circuitry not required for a desired precision level may be switched off.Type: ApplicationFiled: November 24, 2008Publication date: May 27, 2010Applicant: NVIDIA CorporationInventors: Shashank Garg, Vinayak Jayaram Pore, Soumenkumar Dey, Manish Jatashanker Pandey, Harikrishna Madadi Reddy, Manindra Nath Parhy
-
Publication number: 20100103281Abstract: Multiple sets of pixel values representing a captured image of a scene are received, with each set representing an image captured with a corresponding degree of focus. An image processor may identify a region of interest in the captured image, automatically determine the configuration parameters for a lens assembly to provide a desired degree of focus for the region of interest, and generate signals to configure a lens assembly. In an embodiment, the region of interest is a face, the desired degree of focus of the face is determined by computing a rate of variation of luminance of pixels representing the face, and the desired degree is the degree of the image having the maximum degree of focus.Type: ApplicationFiled: December 30, 2009Publication date: April 29, 2010Applicant: NVIDIA CORPORATIONInventors: Shashank Garg, Vinayak Pore, Soumenkumar Dey
-
Publication number: 20080226278Abstract: Multiple sets of pixel values representing a captured image of a scene are received, with each set representing an image captured with a corresponding degree of focus. An image processor may identify a region of interest in the captured image, automatically determine the configuration parameters for a lens assembly to provide a desired degree of focus for the region of interest, and generate signals to configure a lens assembly. In an embodiment, the region of interest is a face, the desired degree of focus of the face is determined by computing a rate of variation of luminance of pixels representing the face, and the desired degree is the degree of the image having the maximum degree of focus.Type: ApplicationFiled: March 15, 2007Publication date: September 18, 2008Applicant: NVIDIA CorporationInventors: Shashank Garg, Vinayak Pore, Soumenkumar Dey
-
Publication number: 20080226279Abstract: An image processor, which determines appropriate exposure parameters for a shutter assembly in a camera. The image processor may computationally determine a region of interest in a scene sought to be captured, and set the parameters to ensure that the exposure parameters are set to capture an image of the scene with the region of interest having a desired brightness level. In an embodiment, pixel values of multiple frames (each frame with a corresponding set of configuration parameters of the shutter assembly) may be examined to determine the frame having pixel values with the region having the desired brightness level. The shutter assembly may be configured with the parameters corresponding to such a frame to provide an auto-exposure feature.Type: ApplicationFiled: March 15, 2007Publication date: September 18, 2008Applicant: NVIDIA CorporationInventors: Soumenkumar Dey, Vinayak Pore, Jaiprakash Nagendra Rao, Ankur Gupta
-
Publication number: 20080225944Abstract: A technique of encoding video frames allocates an available number of bits to different portions of the video frame. A processing unit identifies a region of interest (ROI) in a video frame, and computes a first and second complexity parameter respectively representing the change in video information in the ROI portions and non-ROI portions in the video frame relative to a reference frame. Bits are allocated to the ROI portion proportional (positive correlation) to the first complexity parameter and a ratio of the area of the ROI to the area of the frame. The remaining available bits are allocated to the non-ROI. In an embodiment, the bits are encoded according to H.264 standard.Type: ApplicationFiled: March 15, 2007Publication date: September 18, 2008Applicant: NVIDIA CorporationInventors: Vinayak Pore, Soumenkumar Dey, Vikas Patel