View source for 18500672. Quantization at Different Levels for Data Used in Artificial Neural Network Computations simplified abstract (Micron Technology, Inc.)

=Quantization at Different Levels for Data Used in Artificial Neural Network Computations=


==Organization Name==

[[:Category:Micron Technology, Inc.|Micron Technology, Inc.]]


[[Category:Micron Technology, Inc.]]

==Inventor(s)==

[[:Category:Saideep Tiku of Folsom CA (US)|Saideep Tiku of Folsom CA (US)]][[Category:Saideep Tiku of Folsom CA (US)]]

[[:Category:Shashank Bangalore Lakshman of Folsom CA (US)|Shashank Bangalore Lakshman of Folsom CA (US)]][[Category:Shashank Bangalore Lakshman of Folsom CA (US)]]

[[:Category:Poorna Kale of Folsom CA (US)|Poorna Kale of Folsom CA (US)]][[Category:Poorna Kale of Folsom CA (US)]]

==Quantization at Different Levels for Data Used in Artificial Neural Network Computations - A simplified explanation of the abstract==


This abstract first appeared for US patent application 18500672 titled 'Quantization at Different Levels for Data Used in Artificial Neural Network Computations

==Simplified Explanation==

The patent application describes a pair of smart glasses with a digital camera and a processing device that uses an artificial neural network to analyze images and present virtual content superimposed on the user's view of reality.

* The smart glasses have a digital camera to capture images and a processing device to analyze the images using an artificial neural network with weight data.
* The processing device can apply different quantization levels to data from different regions of the image and apply these levels to the weight data for weighing the data from different regions.
* The glasses can present virtual content based on the output of the artificial neural network superimposed on the user's view.

== Potential Applications ==
The technology can be used in augmented reality applications, healthcare for medical imaging analysis, security for surveillance and facial recognition, and gaming for interactive experiences.

== Problems Solved ==
This technology solves the problem of reducing energy consumption by applying different levels of accuracy to data from different regions of an image during analysis.

== Benefits ==
The benefits of this technology include improved efficiency in image analysis, enhanced user experience with virtual content, and reduced energy consumption in smart glasses.

== Potential Commercial Applications ==
Commercial applications of this technology include augmented reality devices for consumers, medical imaging devices for healthcare professionals, surveillance systems for security companies, and gaming devices for entertainment companies.

== Possible Prior Art ==
One possible prior art could be the use of artificial neural networks in image analysis and augmented reality applications in smart glasses.

=== Unanswered Questions ===

== How does the technology handle privacy concerns related to capturing and analyzing images through the smart glasses? ==
The technology does not address privacy concerns related to capturing and analyzing images through the smart glasses. This aspect would need to be further developed to ensure user privacy and data security.

== What are the potential limitations or challenges in implementing this technology on a larger scale? ==
The technology may face challenges in terms of processing power, data storage, and connectivity requirements for real-time image analysis and presentation of virtual content. Additionally, user acceptance and adoption of smart glasses with this technology may also pose challenges in the market.


==Original Abstract Submitted==

A pair of smart glasses having: a digital camera configured to capture an image of a field of view; and a processing device configured to perform an analysis of the image using an artificial neural network having weight data. The processing device can apply different quantization levels to data from different regions of the image, and apply the different quantization levels to the weight data in weighing on the data from the different regions respectively. For example, weighing image data from a peripheral region of the image with the weight data can be performed with a lower level of accuracy than weighing image data from a center region of the image with the weight data to reduce energy consumption. Based on an output of the artificial neural network responsive to the image, the glasses can present virtual content superimposed on a view of reality seen through the glasses.



[[Category:G06V10/82]]
[[Category:G06F7/544]]
[[Category:G06T11/00]]
[[Category:G06V10/22]]
[[Category:G06V10/24]]
[[Category:G06V10/94]]
[[Category:G06V20/20]]