US 12,395,526 B2
Detecting identity spoofing attacks in multi-sensor systems and applications
Manoj Kumar Yennapureddy, Apex, NC (US); Shagan Sah, Santa Clara, CA (US); and Rajath Shetty, Sunnyvale, CA (US)
Assigned to NVIDIA Corporation, Santa Clara, CA (US)
Filed by NVIDIA Corporation, Santa Clara, CA (US)
Filed on Jul. 12, 2022, as Appl. No. 17/863,140.
Prior Publication US 2024/0022601 A1, Jan. 18, 2024
Int. Cl. H04L 9/40 (2022.01); G06T 7/50 (2017.01)
CPC H04L 63/1466 (2013.01) [G06T 7/50 (2017.01); H04L 63/1416 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/20076 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, using one or more image sensors located at one or more first locations within a vehicle, image data representative of one or more images depicting a user;
receiving, using one or more depth sensors located at one or more second locations within the vehicle, depth data representative of one or more distance values associated with the user located within the vehicle;
generating, based at least on one or more first layers of one or more neural networks processing the image data, one or more first feature maps corresponding to one or more first features as represented by the image data;
generating, based at least on one or more second layers of the one or more neural networks processing the depth data, one or more second feature maps corresponding to one or more second features represented by the depth data;
generating, based at least on the one or more neural networks fusing the one or more first feature maps with the one or more second feature maps, one or more fused feature maps;
generating, based at least on the one or more neural networks processing the one or more fused feature maps, output data indicating whether a spoofing attack is occurring with regard to the user; and
performing one or more operations based at least on whether the spoofing attack is occurring with regard to the user.