US 11,748,998 B1
Three-dimensional object estimation using two-dimensional annotations
Hanlin Goh, Sunnyvale, CA (US); Nitish Srivastava, San Francisco, CA (US); Yichuan Tang, Santa Clara, CA (US); and Ruslan Salakhutdinov, Pittsburgh, PA (US)
Assigned to APPLE INC., Cupertino, CA (US)
Filed by Hanlin Goh, Sunnyvale, CA (US); Nitish Srivastava, San Francisco, CA (US); Yichuan Tang, Santa Clara, CA (US); and Ruslan Salakhutdinov, Pittsburgh, PA (US)
Filed on May 25, 2022, as Appl. No. 17/824,202.
Application 17/824,202 is a continuation of application No. 16/433,213, filed on Jun. 6, 2019, granted, now 11,373,411.
Claims priority of provisional application 62/738,397, filed on Sep. 28, 2018.
Claims priority of provisional application 62/684,248, filed on Jun. 13, 2018.
Int. Cl. G06V 20/58 (2022.01); G06N 20/00 (2019.01); G06T 7/33 (2017.01); B60W 40/04 (2006.01); G06V 10/75 (2022.01); G06T 15/20 (2011.01)
CPC G06V 20/58 (2022.01) [B60W 40/04 (2013.01); G06N 20/00 (2019.01); G06T 7/337 (2017.01); G06T 15/205 (2013.01); G06V 10/757 (2022.01)] 21 Claims
OG exemplary drawing
 
1. A non-transitory computer-readable storage device including program instructions executable by one or more processors that, when executed, cause the one or more processors to perform operations, the operations comprising:
obtaining a two-dimensional image;
obtaining a two-dimensional image annotation that indicates presence of an object in the two-dimensional image;
obtaining three-dimensional sensor information;
generating a top-down representation of the three-dimensional sensor information;
obtaining a top-down annotation that indicates presence of the object in the top-down representation;
determining a bottom surface of a three-dimensional cuboid based on map information;
determining a position, a length, a width, and a yaw rotation of the three-dimensional cuboid based on the top-down annotation; and
determining a height of the three-dimensional cuboid based on the two-dimensional image annotation, and the position, the length, the width, and the yaw rotation of the three-dimensional cuboid.