US 11,810,064 B2
Method(s) and system(s) for vehicular cargo management
Amruta Kulkarni, Santa Clara, CA (US); Naveen Sangeneni, Fremont, CA (US); Kevin Abas, Fremont, CA (US); Reddy Avula Srinivas, Fremont, CA (US); Sreenjoy Chatterjee, San Francisco, CA (US); Andrew Kaneshiro, Los Altos, CA (US); Jyh-Yung Lin, Santa Clara, CA (US); and Gagandeep Singh, San Jose, CA (US)
Assigned to Mercedes-Benz Group AG, Stuttgart (DE)
Appl. No. 17/271,900
Filed by Mercedes-Benz Group AG, Stuttgart (DE)
PCT Filed Jun. 27, 2019, PCT No. PCT/EP2019/067259
§ 371(c)(1), (2) Date Feb. 26, 2021,
PCT Pub. No. WO2020/043351, PCT Pub. Date Mar. 5, 2020.
Claims priority of application No. 10 2018 006 764.6 (DE), filed on Aug. 27, 2018.
Prior Publication US 2021/0192780 A1, Jun. 24, 2021
Int. Cl. G06Q 10/087 (2023.01); G06T 7/73 (2017.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01); G06V 20/59 (2022.01); G06F 18/23 (2023.01); G06Q 10/0631 (2023.01); G06Q 50/28 (2012.01)
CPC G06Q 10/087 (2013.01) [G06F 18/23 (2023.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01); G06T 7/73 (2017.01); G06V 20/59 (2022.01); G06Q 10/06315 (2013.01); G06Q 50/28 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30268 (2013.01)] 5 Claims
OG exemplary drawing
 
1. A method for identifying a position of an object in a cargo space, comprising the steps of:
identifying a region of interest (ROI) from a field of view of the cargo space being captured by one or more cameras, wherein the one or more cameras capture at least one of depth and color of the field of view;
extracting a plurality of planes from the ROI, wherein the plurality of planes correspond to a geometry of the object;
clustering similar and nearby planes, wherein the clustering is based on a weight of two or more planes of the plurality of planes, and wherein the weight is assigned based on a property of orthogonality and a property of dimensions of two or more planes of the plurality of planes;
modelling a multi-dimensional bounding box corresponding to the object based on the clustered planes; and
identifying a position of the object based on a position of the multi-dimensional bounding box, wherein the weight assigned based on the property of orthogonality is more than the weight assigned based on the property of dimensions of the two or more planes of the plurality of planes.