US 11,741,710 B2
Accelerated video processing for feature recognition via an artificial neural network configured in a data storage device
Poorna Kale, Folsom, CA (US)
Assigned to Micron Technology, Inc., Boise, ID (US)
Filed by Micron Technology, Inc., Boise, ID (US)
Filed on Oct. 22, 2020, as Appl. No. 17/77,981.
Prior Publication US 2022/0129677 A1, Apr. 28, 2022
Int. Cl. G06V 20/40 (2022.01); G06N 3/08 (2023.01)
CPC G06V 20/40 (2022.01) [G06N 3/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving first data representative of a first video frame in a video stream;
determining from the first data a first feature identification representative of one or more features captured in the first video frame;
receiving second data representative of a second video frame in the video stream;
generating an intermediate result of analysis of the second data using a first portion of an artificial neural network, the intermediate result identifying a preliminary feature identification of one or more features captured in the second video frame and a confidence level of the preliminary feature identification;
making a determination that the confidence level of the preliminary feature identification for the second video frame is above a threshold based at least in part on the first feature identification for the first video frame; and
terminating, responsive to the determination, computation of a second portion of the artificial neural network.