- Santa Clara CA, US Mini Rawat - San Jose CA, US Tao Li - San Jose CA, US Divya Ramakrishnan - San Jose CA, US Swanand Santosh Bindoo - San Jose CA, US Sreedhar Narayanaswamy - Sunnyvale CA, US
International Classification:
H05K 7/14 G06T 1/20 G06F 9/50
Abstract:
A system to select graphics processing units (GPUs) to execute a task is disclosed. In at least one embodiment, GPUs are selected based on one or more task parameters and one or more fused parameters.
Estimating Tracking Sensor Parametrization Using Known Surface Constraints
- Durham NC, US Divya Ramakrishnan - Santa Clara CA, US Lavanya Sridharan - Fremont CA, US Richard Cavallaro - Mountain View CA, US Marvin S. WHITE - San Carlos CA, US
A sensor system and a method of operating a sensor system including a plurality of sensors tracking a moving object in an area having known bounding surfaces. The apparatus and method calculate a time-specific position of the object based on data and sensor parameters from at least two of the plurality of sensors and determine errors between the calculated time-specific positions calculated. The method and apparatus calculate a minimum system error attributable to the at least two sensors by constraining at least one dimension in the data of the sensor used in the calculated time-specific position of the object associated with the sensor, the constraining based on an object/surface interaction, the minimum system error calculated by solving for modified sensor parameters for each sensor.
- Durham NC, US Radford Parker - San Mateo CA, US Divya Ramakrishnan - Santa Clara CA, US Louis Gentry - Union City CA, US Rand Pendleton - Santa Cruz CA, US
Assignee:
SportsMEDIA Technology Corporation - Durham NC
International Classification:
G06T 7/80 G06T 7/73 G06T 7/30
Abstract:
A system for registering one or more cameras and/or creating an accurate three-dimensional (3D) model of a world space environment including back projecting at least one image from at least one of a plurality of camera views to the 3D model based on a set of existing camera parameters. The back projected image is added as a texture for the 3D model. This texture is automatically compared to one or more images from other camera views using a color space comparison of images to determine a set of differences or errors. The camera parameters and the 3D model are automatically adjusted to minimized the differences or errors. Over time, the parameters and the 3D model converge on a state that can be used to track moving objects, insert virtual graphics and/or perform other functions.
Dynamic Seam Adjustment Of Image Overlap Zones From Multi-Camera Source Images
- San Jose CA, US Vidya Elangovan - Cupertino CA, US Divya Ramakrishnan - Santa Clara CA, US Prashant Jain - San Jose CA, US
International Classification:
H04N 5/232 H04N 7/18 G06K 9/00
Abstract:
Embodiments of the present disclosure are directed to providing a more accurate presentation of a vehicle's surroundings when the presentation combines images from multiple cameras. Cameras may capture the same portion of the environment. If an object is within view of at least two cameras, each camera produces a portion of an overall view, such as a 360 degree synthetic top-down view. The seam, angle and/or geometry, angle between any two images is dynamically determined, such as due to proximity (or lack thereof) to an object. As a result the synthetic top-down view may present an image of an object more prominently and/or avoid having the seam between camera images fall on the image of the object, which may otherwise result in the image of the object being omitted from both camera images.
Estimating Tracking Sensor Parametrization Using Known Surface Constraints
- Durham NC, US Divya Ramakrishnan - Santa Clara CA, US Lavanya Sridharan - Fremont CA, US Richard Cavallaro - Mountain View CA, US Marvin S. WHITE - San Carlos CA, US
A sensor system and a method of operating a sensor system including a plurality of sensors tracking a moving object in an area having known bounding surfaces. The apparatus and method calculate a time-specific position of the object based on data and sensor parameters from at least two of the plurality of sensors and determine errors between the calculated time-specific positions calculated. The method and apparatus calculate a minimum system error attributable to the at least two sensors by constraining at least one dimension in the data of the sensor used in the calculated time-specific position of the object associated with the sensor, the constraining based on an object/surface interaction, the minimum system error calculated by solving for modified sensor parameters for each sensor.
Techniques For Configuring A Processor To Execute Instructions Efficiently
- Santa Clara CA, US Tezaswi RAJA - San Jose CA, US Kevin WILDER - Menlo Park CA, US Raj SELVANESAN - Cupertino CA, US Divya RAMAKRISHNAN - San Jose CA, US Daniel RODRIGUEZ - Dublin CA, US Benjamin FAULKNER - San Jose CA, US Raj JAYAKAR - San Jose CA, US Walter LI - Milpitas CA, US
International Classification:
G06F 1/324 G06F 9/38
Abstract:
Systems and techniques for improving the performance of circuits while adapting to dynamic voltage drops caused by the execution of noisy instructions (e.g. high power consuming instructions) are provided. The performance is improved by slowing down the frequency of operation selectively for types of noisy instructions. An example technique controls a clock by detecting an instruction of a predetermined noisy type that is predicted to have a predefined noise characteristic (e.g. a high level of noise generated on the voltage rails of a circuit due to greater amount of current drawn by the instruction), and, responsive to the detecting, deceasing a frequency of the clock. The detecting occurs before execution of the instruction. The changing of the frequency in accordance with instruction type enables the circuits to be operated at high frequencies even if some of the workloads include instructions for which the frequency of operation is slowed down.
- Durham NC, US Radford Parker - San Mateo CA, US Divya Ramakrishnan - Santa Clara CA, US Louis Gentry - Union City CA, US Rand Pendleton - Santa Cruz CA, US
Assignee:
SportsMEDIA Technology Corporation - Durham NC
International Classification:
G06T 7/80 G06T 7/30 G06T 7/73
Abstract:
A system for registering one or more cameras and/or creating an accurate three-dimensional (3D) model of a world space environment including back projecting at least one image from at least one of a plurality of camera views to the 3D model based on a set of existing camera parameters. The back projected image is added as a texture for the 3D model. This texture is automatically compared to one or more images from other camera views using a color space comparison of images to determine a set of differences or errors. The camera parameters and the 3D model are automatically adjusted to minimized the differences or errors. Over time, the parameters and the 3D model converge on a state that can be used to track moving objects, insert virtual graphics and/or perform other functions.
Estimating Tracking Sensor Parametrization Using Known Surface Constraints
- Durham NC, US Divya Ramakrishnan - Santa Clara CA, US Lavanya Sridharan - Fremont CA, US Richard H. CAVALLARO - Mountain View CA, US Marvin S. WHITE - San Carlos CA, US
A sensor system and a method of operating a sensor system including a plurality of sensors tracking a moving object in an area having known bounding surfaces. The apparatus and method calculate a time-specific position of the object based on data and sensor parameters from at least two of the plurality of sensors and determine errors between the calculated time-specific positions calculated. The method and apparatus calculate a minimum system error attributable to the at least two sensors by constraining at least one dimension in the data of the sensor used in the calculated time-specific position of the object associated with the sensor, the constraining based on an object/surface interaction, the minimum system error calculated by solving for modified sensor parameters for each sensor.
Youtube
Divya Ramakrishnan - Winner (Impact Top 30 Un...
exchange4media is a single stop information platform for the entire in...