US 11,854,228 B2
Methods and systems for volumetric modeling independent of depth data
Tom Hsi Hao Shang, Forest Hills, NY (US); Elena Dotsenko, Princeton, NJ (US); and Liang Luo, Nutley, NJ (US)
Assigned to Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed by Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed on Apr. 5, 2022, as Appl. No. 17/713,914.
Application 17/713,914 is a continuation of application No. 17/073,108, filed on Oct. 16, 2020, granted, now 11,328,445.
Prior Publication US 2022/0230356 A1, Jul. 21, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/73 (2017.01); G06T 17/00 (2006.01); G06T 7/80 (2017.01); H04N 23/90 (2023.01)
CPC G06T 7/74 (2017.01) [G06T 7/80 (2017.01); G06T 17/00 (2013.01); H04N 23/90 (2023.01); G06T 2200/08 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30196 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
determining, by an image processing system, that an object partially depicted in an image is an instance of an object type for which a machine learning model is available to the image processing system;
obtaining, by the image processing system in response to the determining that the object is the instance of the object type, pose data generated based on the machine learning model to provide insight about pose capabilities of the object type by representing how objects of the object type are capable of being posed; and
generating, by the image processing system, a full volumetric representation of the object in a pose that is estimated, independently of depth data for the object, based on the image and based on the insight about the pose capabilities of the object type provided by the pose data.