US 11,790,648 B2
Automated usability assessment of buildings using visual data of captured in-room images
Viktoriya Stoeva, Phoenix, AZ (US); Sing Bing Kang, Redmond, WA (US); Naji Khosravan, Seattle, WA (US); and Lambert E. Wixson, Bellevue, WA (US)
Assigned to MFTB Holdco, Inc., Seattle, WA (US)
Filed by MFTB Holdco, Inc., Seattle, WA (US)
Filed on Feb. 25, 2021, as Appl. No. 17/185,793.
Prior Publication US 2022/0269888 A1, Aug. 25, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 20/20 (2022.01); G06T 7/55 (2017.01); G06N 3/08 (2023.01); G06T 19/00 (2011.01)
CPC G06V 20/20 (2022.01) [G06N 3/08 (2013.01); G06T 7/55 (2017.01); G06T 19/006 (2013.01); G06T 2210/04 (2013.01); G06T 2210/12 (2013.01); G06T 2210/61 (2013.01)] 30 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
obtaining, by one or more computing systems, one or more panorama images captured in a room of a house and having visual data that in combination provide 360 degrees of horizontal visual coverage of the room;
analyzing, by the one or more computing systems and via use of at least a first trained neural network, the visual data of the one or more panorama images to identify multiple objects installed in the room and to assess a layout of the room with respect to an indicated purpose of the room;
determining, by the one or more computing systems and for each of the multiple objects, one or more target attributes of that object for which to capture additional data based at least in part on the visual data lacking details to satisfy a defined threshold about the one or more target attributes of that object;
providing, by the one or more computing systems, instructions to capture the additional data about the one or more target attributes of each of the multiple objects, wherein capturing of the additional data includes obtaining additional perspective images of one or more specified types of the multiple objects;
analyzing, by the one or more computing systems and via use of at least a second trained neural network, additional visual data of the additional perspective images to, for each of the multiple objects, verify that the additional data about the one or more target attributes of that object have been captured and to generate an assessment, based at least in part on the one or more target attributes of that object, of a current contribution of that object to usability of the room for the indicated purpose;
determining, by the one or more computing systems, and based at least in part on combining information about the assessed layout of the room and the assessments of the current contributions of the multiple objects, an assessment of the current usability of the room for the indicated purpose; and
displaying, by the one or more computing systems, information about the determined assessment of the current usability of the room along with additional visual information about the room.