US 12,318,951 B2
Systems and methods for object detection
Ahmed Abouelela, Tokyo (JP); Hamdi Sahloul, Tokyo (JP); Jose Jeronimo Moreira Rodrigues, Tokyo (JP); Xutao Ye, Tokyo (JP); and Jinze Yu, Tokyo (JP)
Filed by MUJIN, INC., Tokyo (JP)
Filed on Aug. 9, 2022, as Appl. No. 17/884,151.
Claims priority of provisional application 63/230,931, filed on Aug. 9, 2021.
Prior Publication US 2023/0044001 A1, Feb. 9, 2023
Int. Cl. G06T 7/70 (2017.01); B25J 9/16 (2006.01); G06T 7/00 (2017.01); G06T 7/73 (2017.01); G06V 10/44 (2022.01); G06V 10/75 (2022.01); G06V 10/77 (2022.01); G06V 10/776 (2022.01); G06V 20/50 (2022.01); G06V 20/64 (2022.01)
CPC B25J 9/1697 (2013.01) [B25J 9/1664 (2013.01); B25J 9/1669 (2013.01); G06T 7/0002 (2013.01); G06T 7/70 (2017.01); G06T 7/74 (2017.01); G06V 10/443 (2022.01); G06V 10/751 (2022.01); G06V 10/7515 (2022.01); G06V 10/7715 (2022.01); G06V 10/776 (2022.01); G06V 20/50 (2022.01); G06V 20/653 (2022.01); G06T 2207/30168 (2013.01); G06V 2201/06 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computing system comprising:
at least one processing circuit in communication with a robot, having an arm and an end-effector connected thereto, and a camera having a field of view and configured, when one or more objects are or have been in the field of view, to execute instructions stored on a non-transitory computer-readable medium for:
obtaining object image information of an object in a scene;
generating one or more detection hypotheses, wherein the one or more detection hypotheses correspond to an object recognition template representing a template object; and
validating each detection hypothesis of the one or more detection hypotheses by:
generating one or more three-dimensional validation scores based on comparing three-dimensional information of the object recognition template of the one or more detection hypotheses and three-dimensional information of the object image information corresponding to the object, the one or more three-dimensional validation scores including at least one of an occlusion validator score, a three-demensional information validator score, a hole matching validator score, and a normal vector validator score;
generating one or more two-dimensional validation scores based on comparing two-dimensional information of the corresponding object recognition template of the one or more detection hypotheses and two-dimensional information of the object image information, the one or more two-dimensional validation scores including at least one of a rendered match validator score and a template match validator score;
selecting a detection hypothesis from the one or more detection hypotheses according to the one or more three-dimensional validation scores and the one or more two-dimensional validation scores; and
generating a detection result for the object in the scene according to the selected detection hypothesis.