US 11,694,362 B2
Data processing systems for real-time camera parameter estimation
Leonardo Citraro, S. Antonio (CH); Pablo Márquez Neila, Lausanne (CH); Stefano Savarè, Lausanne (CH); Vivek Jayaram, Los Gatos, CA (US); Charles Xavier Quentin Dubout, Écublens (CH); Felix Constant Marc Renaut, Morges (CH); Andres Michael Levering Hasfura, San Antonio, TX (US); Horesh Beny Ben Shitrit, Echichens (CH); and Pascal Fua, Vaux sur Morges (CH)
Assigned to Genius Sports SS, LLC, Los Angeles, CA (US)
Filed by Genius Sports SS, LLC, Los Angeles, CA (US)
Filed on Apr. 7, 2021, as Appl. No. 17/224,207.
Application 17/224,207 is a continuation of application No. 16/798,900, filed on Feb. 24, 2020, granted, now 10,991,125.
Application 16/798,900 is a continuation of application No. 16/521,761, filed on Jul. 25, 2019, granted, now 10,600,210, issued on Mar. 24, 2020.
Prior Publication US 2021/0225034 A1, Jul. 22, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/08 (2023.01); G06T 7/80 (2017.01); G06V 20/40 (2022.01)
CPC G06T 7/80 (2017.01) [G06N 3/08 (2013.01); G06V 20/42 (2022.01); G06V 20/46 (2022.01); G06V 20/48 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A system for estimating one or more camera parameters, the system comprising:
one or more computer processors; and
memory storing computer-executable instructions that, when executed by the one or more computer processors, cause the one or more computer processors to perform operations comprising:
determining one or more semantic keypoints for an image in an image sequence;
determining one or more person keypoints for the image;
generating a first estimated homography using the one or more semantic keypoints for the image;
projecting each of the one or more person keypoints for the image onto a world coordinate system of a particular environment;
determining, for each of the one or more person keypoints for the image, based at least in part on projecting each of the one or more person keypoints for the image onto the world coordinate system of the particular environment, a proximate person keypoint location in the world coordinate system of the particular environment;
determining, for each of the one or more semantic keypoints for the image, a semantic keypoint location in the world coordinate system of the particular environment;
generating a second estimated homography using the one or more semantic keypoints for the image, the semantic keypoint location for each of the one or more semantic keypoints for the image, the one or more person keypoints for the image, and the proximate person keypoint location for each of the one or more person keypoints for the image;
determining a final estimated homography based at least in part on the second estimated homography;
determining one or more intrinsic parameter estimates based at least in part on the final estimated homography; and
determining one or more extrinsic parameter estimates based at least in part on the final estimated homography.