FRAGMENTED GRAPHIC CORES FOR DEEP LEARNING USING LED DISPLAYS
First Claim
Patent Images
1. A smart display, comprising:
- one or more groups of smart pixels; and
at least one graphics engine, wherein the at least one graphics engine is fragmented into GPU (graphics processing unit) minute cores, the GPU minute cores being distributed throughout the smart display;
wherein smart pixels with distributed graphics within the smart display perform deep learning.
1 Assignment
0 Petitions
Accused Products
Abstract
A smart display including one or more groups of smart pixels and at least one graphics engine. The at least one graphics engine is fragmented into GPU (graphics processing unit) minute cores. The GPU minute cores are distributed throughout the smart display. The smart pixels with distributed graphics within the smart display perform deep learning. Libraries stored on GPU minute core embedded memory are used to perform object identification using deep learning. The smart display monitors for pixel degradation and, if necessary, performs pixel enhancement.
-
Citations
25 Claims
-
1. A smart display, comprising:
-
one or more groups of smart pixels; and at least one graphics engine, wherein the at least one graphics engine is fragmented into GPU (graphics processing unit) minute cores, the GPU minute cores being distributed throughout the smart display; wherein smart pixels with distributed graphics within the smart display perform deep learning. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A method for object identification, comprising:
-
displaying an object to be identified on smart pixels of a smart display; scanning, by at least one GPU (graphics processing unit) minute core, a library having a subject matter identified by the GPU minute core as being the subject matter of the displayed object; if the object is found in the library, identifying the object by name on the smart display; performing, by the GPU minute core, deep learning techniques to predict something about the object in anticipation of the needs of the user of the smart display; and updating the library with the information obtained through identification and deep learning. - View Dependent Claims (19, 20)
-
-
21. At least one computer readable medium, comprising a set of instructions, which when executed by at least one computing device cause the at least one computing device to:
-
display an object to be identified on smart pixels of a smart display; scan, by at least one GPU (graphics processing unit) minute core of the smart display, a library having a subject matter identified by the GPU minute core as being the subject matter of the displayed object; if the object is found in the library, identify the object by name on the smart display; perform, by the GPU minute core, deep learning techniques to predict something about the object in anticipation of the needs of the user of the smart display; and update the library with the information obtained through identification and deep learning. - View Dependent Claims (22, 23)
-
-
24. A method for pixel enhancement on a smart display, comprising:
-
monitoring, by the smart display, the brightness of the smart pixels on the smart display; determining if there is a region of the smart display in which the brightness of the smart pixels is degrading; if there is a region in which the brightness of the smart pixels is degrading, determining if the brightness level of the smart pixels that are degrading has fallen below a pre-defined threshold; if the brightness level has fallen below a pre-defined threshold, determining if the region is an area in which the smart pixels can be turned off; if the pixels can be turned off, turning the degraded smart pixels off to relax the smart display in the affected area; if the pixels cannot be turned off, applying additional current and voltage to the pixels in the affected region to restore the smart pixels to an acceptable brightness level. - View Dependent Claims (25)
-
Specification