US 11,989,900 B2
Object recognition neural network for amodal center prediction
Siddharth Mahendran, San Jose, CA (US); Nitin Bansal, Sunnyvale, CA (US); Nitesh Sekhar, Mountain View, CA (US); Manushree Gangwar, San Francisco, CA (US); Khushi Gupta, Mountain View, CA (US); and Prateek Singhal, Mountain View, CA (US)
Assigned to Magic Leap, Inc., Plantation, FL (US)
Filed by Magic Leap, Inc., Plantation, FL (US)
Filed on Jun. 24, 2021, as Appl. No. 17/357,118.
Claims priority of provisional application 63/043,463, filed on Jun. 24, 2020.
Prior Publication US 2021/0407125 A1, Dec. 30, 2021
Int. Cl. G06T 7/73 (2017.01); G06F 18/21 (2023.01); G06F 18/24 (2023.01); G06N 3/04 (2023.01); G06T 7/579 (2017.01); G06T 7/60 (2017.01); G06T 19/00 (2011.01); G06V 10/44 (2022.01); G06V 20/20 (2022.01); G06V 20/64 (2022.01)
CPC G06T 7/579 (2017.01) [G06F 18/2163 (2023.01); G06F 18/24 (2023.01); G06N 3/04 (2013.01); G06T 7/60 (2013.01); G06T 7/73 (2017.01); G06T 19/006 (2013.01); G06V 10/454 (2022.01); G06V 20/20 (2022.01); G06V 20/647 (2022.01); G06T 2207/20084 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method, the method comprising:
receiving an image of an object captured by a camera;
processing the image of the object using an object recognition neural network that is configured to generate an object recognition output comprising:
data defining a predicted two-dimensional amodal center of the object, wherein the predicted two-dimensional amodal center of the object is a projection of a predicted three-dimensional center of the object under a camera pose of the camera that captured the image;
obtaining data specifying one or more other predicted two-dimensional amodal centers of the object in one or more other images captured under different camera poses; and
determining, from (i) the predicted two-dimensional amodal center of the object in the image and (ii) the one or more other predicted two-dimensional amodal centers of the object, the predicted three-dimensional center of the object.