US 11,869,118 B2
Generating a synthetic ground-truth image using a dead leaves model
Pavan Chennagiri, Austin, TX (US); John Seokjun Lee, Allen, TX (US); and Hamid R. Sheikh, Allen, TX (US)
Assigned to Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed by Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed on Jan. 28, 2022, as Appl. No. 17/588,024.
Claims priority of provisional application 63/247,117, filed on Sep. 22, 2021.
Prior Publication US 2023/0091909 A1, Mar. 23, 2023
Int. Cl. G06T 11/00 (2006.01); G06T 19/20 (2011.01); G06T 7/90 (2017.01); G06T 5/00 (2006.01); G06N 20/00 (2019.01); G06T 17/00 (2006.01)
CPC G06T 11/001 (2013.01) [G06N 20/00 (2019.01); G06T 5/002 (2013.01); G06T 7/90 (2017.01); G06T 17/00 (2013.01); G06T 19/20 (2013.01); G06T 2207/10012 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/20081 (2013.01); G06T 2219/2012 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus comprising:
at least one memory configured to store an artificial intelligence (AI) network; and
at least one processor configured to:
generate a dead leaves model;
capture a ground truth frame from the dead leaves model;
apply a mathematical noise model to the ground truth frame to produce a noisy frame; and
train the AI network using the ground truth frame and the noisy frame;
wherein, to capture the ground truth frame from the dead leaves model, the at least one processor is configured to generate a dead leaves object and apply a texture to the dead leaves object.