US 12,337,478 B2
Method, computer system, and non-transitory computer-readable record medium for learning robot skill
Taeyoon Lee, Seongnam-si (KR); Choongin Lee, Seongnam-si (KR); Changwoo Park, Seongnam-si (KR); Keunjun Choi, Seongnam-si (KR); Donghyun Sung, Seongnam-si (KR); Kyoungyeon Choi, Seongnam-si (KR); Younghyo Park, Seongnam-si (KR); and Seunghun Jeon, Seongnam-si (KR)
Assigned to Naver Corporation, Gyeonggi-do (KR)
Filed by NAVER CORPORATION, Seongnam-si (KR)
Filed on Sep. 13, 2024, as Appl. No. 18/884,894.
Application 18/884,894 is a continuation of application No. PCT/KR2023/002997, filed on Mar. 6, 2023.
Claims priority of application No. 10-2022-0031883 (KR), filed on Mar. 15, 2022; and application No. 10-2022-0044696 (KR), filed on Apr. 11, 2022.
Prior Publication US 2025/0010466 A1, Jan. 9, 2025
Int. Cl. B25J 9/16 (2006.01); B25J 11/00 (2006.01)
CPC B25J 9/163 (2013.01) [B25J 9/1661 (2013.01); B25J 9/1669 (2013.01); B25J 9/1679 (2013.01); B25J 9/1697 (2013.01); B25J 11/0075 (2013.01); B25J 9/161 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A robotic painting method performed by a computer system, wherein the computer system comprises at least one processor configured to execute computer-readable instructions included in a memory, and
the robotic painting method comprises:
collecting, by the at least one processor, stroke-level action data for a painting action;
learning, by the at least one processor, a stroke-level robotic painting skill using the stroke-level action data, wherein the learning includes training a deep latent variable model using the stroke-level action data; and
performing, by the at least one processor, a painting task for a given target image based on a prediction using the deep latent variable model.