US 11,941,819 B2
Object detection using skewed polygons suitable for parking space detection
Dongwoo Lee, Seoul (KR); Junghyun Kwon, San Jose, CA (US); Sangmin Oh, San Jose, CA (US); Wenchao Zheng, San Jose, CA (US); Hae-Jong Seo, San Jose, CA (US); David Nister, Bellevue, WA (US); and Berta Rodriguez Hervas, San Francisco, CA (US)
Assigned to NVIDIA Corporation, Santa Clara, CA (US)
Filed by NVIDIA Corporation, Santa Clara, CA (US)
Filed on Dec. 6, 2021, as Appl. No. 17/457,825.
Application 17/457,825 is a continuation of application No. 16/820,164, filed on Mar. 16, 2020, granted, now 11,195,331.
Claims priority of provisional application 62/819,544, filed on Mar. 16, 2019.
Prior Publication US 2022/0092855 A1, Mar. 24, 2022
Int. Cl. G06T 7/13 (2017.01); G06T 7/40 (2017.01); G06T 17/30 (2006.01); G06V 10/44 (2022.01); G06V 10/75 (2022.01); G06V 10/772 (2022.01); G06V 10/82 (2022.01); G06V 20/58 (2022.01)
CPC G06T 7/13 (2017.01) [G06T 7/40 (2013.01); G06T 17/30 (2013.01); G06V 10/454 (2022.01); G06V 10/751 (2022.01); G06V 10/772 (2022.01); G06V 10/82 (2022.01); G06V 20/586 (2022.01); G06T 2207/10021 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30264 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
one or more processing units to execute operations comprising:
applying, to one or more machine learning models (MLMs) trained to predict one or more displacement values to one or more points defining a multi-dimensional shape, sensor data corresponding to at least one field of view of at least one sensor in an environment, the at least one field of view including at least a portion of at least one target space;
receiving, from the one or more MLMs, data indicating the one or more displacement values to the multi-dimensional shape; and
adjusting the multi-dimensional shape using the one or more displacement values to generate a bounding shape corresponding to the at least one target space in one or more images corresponding to the at least one field of view.