US 11,989,822 B2
Damage detection from multi-view visual data
Stefan Johannes Josef Holzer, San Mateo, CA (US); Abhishek Kar, Berkeley, CA (US); Matteo Munaro, San Francisco, CA (US); Pavel Hanchar, Minsk (BY); Radu Bogdan Rusu, San Francisco, CA (US); and Santi Arano, San Francisco, CA (US)
Assigned to Fyusion, Inc., San Francisco, CA (US)
Filed by Fyusion, Inc., San Francisco, CA (US)
Filed on Jun. 20, 2023, as Appl. No. 18/338,240.
Application 18/338,240 is a continuation of application No. 17/174,250, filed on Feb. 11, 2021, granted, now 11,727,626.
Application 17/174,250 is a continuation of application No. 16/692,133, filed on Nov. 22, 2019, granted, now 10,950,033, issued on Mar. 16, 2021.
Claims priority of provisional application 62/795,427, filed on Jan. 22, 2019.
Prior Publication US 2023/0334768 A1, Oct. 19, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 7/70 (2017.01); G01C 21/32 (2006.01); G06F 9/451 (2018.01); G06F 16/29 (2019.01); G06F 17/18 (2006.01); G06F 30/15 (2020.01); G06N 3/02 (2006.01); G06Q 30/02 (2023.01); G06T 7/00 (2017.01); G06T 7/593 (2017.01); G06T 15/10 (2011.01); G06T 15/20 (2011.01); G06T 17/00 (2006.01); G06T 19/00 (2011.01); H04N 13/243 (2018.01); H04N 13/271 (2018.01); H04N 23/63 (2023.01); H04N 13/00 (2018.01)
CPC G06T 15/205 (2013.01) [G01C 21/32 (2013.01); G06F 9/453 (2018.02); G06F 16/29 (2019.01); G06F 17/18 (2013.01); G06F 30/15 (2020.01); G06N 3/02 (2013.01); G06Q 30/0278 (2013.01); G06T 7/0002 (2013.01); G06T 7/0004 (2013.01); G06T 7/593 (2017.01); G06T 7/70 (2017.01); G06T 15/10 (2013.01); G06T 17/00 (2013.01); G06T 19/003 (2013.01); G06T 19/006 (2013.01); H04N 13/243 (2018.05); H04N 13/271 (2018.05); H04N 23/633 (2023.01); G06T 2200/08 (2013.01); G06T 2200/24 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/20076 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20224 (2013.01); G06T 2207/30108 (2013.01); G06T 2207/30244 (2013.01); G06T 2207/30248 (2013.01); H04N 2013/0081 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A method comprising:
determining an object model by applying a neural network to a first plurality of images to create a three-dimensional skeleton of a designated object, each of the first plurality of images being captured from a respective viewpoint, the object model including a plurality of object model components, each of the images corresponding to one or more of the object model components, each of the object model components corresponding to a respective portion of the designated object;
determining respective component condition information for one or more of the object model components using the plurality of images, the component condition information indicating a characteristic of damage incurred by the respective object portion corresponding with the object model component;
converting the characteristic of damage into a heatmap visual representation associated with estimated damage to the respective object portion;
overlaying the heatmap onto a Multi-View Interactive Digital Media Representation (MVIDMR) associated with the designated object; and
storing the component condition information on a storage device.