US 12,450,899 B2
Sensor calibration for space translation
Evan McLaughlin, Guelph (CA); Farzin Aghdasi, East Palo Alto, CA (US); Milind Naphade, Cupertino, CA (US); Arihant Jain, Fremont, CA (US); Sujit Biswas, San Jose, CA (US); and Parthasarathy Sriram, Los Altos Hills, CA (US)
Assigned to Nvidia Corporation, Santa Clara, CA (US)
Filed by Nvidia Corporation, Santa Clara, CA (US)
Filed on May 4, 2021, as Appl. No. 17/307,688.
Claims priority of provisional application 63/020,538, filed on May 5, 2020.
Prior Publication US 2021/0348938 A1, Nov. 11, 2021
Int. Cl. G06V 20/10 (2022.01); G01C 21/34 (2006.01); G01C 21/36 (2006.01); G06V 20/20 (2022.01); G06V 20/54 (2022.01)
CPC G06V 20/182 (2022.01) [G01C 21/3476 (2013.01); G01C 21/3614 (2013.01); G06V 20/20 (2022.01); G06V 20/54 (2022.01)] 9 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
transmitting data causing a presentation of a first image of a scene, taken from a first static perspective corresponding to a first sensor used to obtain the first image;
transmitting data causing a presentation of a second image of the scene, taken from a second perspective corresponding to a second sensor used to obtain the second image;
receiving a plurality of inputs, selected by a user via an interface, corresponding to a plurality of landmarks in the first image that are also depicted in the second image, the plurality of inputs forming a static polygon defining a static calibration region including one or more elevation changes within the static calibration region between the first static perspective and the second perspective;
translating the static polygon, including the plurality of landmarks, to the second image using a homomorphic translation matrix; and calibrating, for the static calibration region defined by the static polygon, the first sensor based on the translated static polygon in the second image.