US 11,941,868 B2
Inference apparatus, inference method, and computer-readable storage medium storing an inference program
Ryo Yonetani, Tokyo (JP); Masaki Suwa, Tokyo (JP); Mohammadamin Barekatain, Tokyo (JP); Yoshihisa Ijiri, Tokyo (JP); and Hiroyuki Miyaura, Kyoto (JP)
Assigned to OMRON CORPORATION, Kyoto (JP)
Appl. No. 17/620,769
Filed by OMRON Corporation, Kyoto (JP)
PCT Filed Jun. 25, 2020, PCT No. PCT/JP2020/024998
§ 371(c)(1), (2) Date Dec. 20, 2021,
PCT Pub. No. WO2021/014878, PCT Pub. Date Jan. 28, 2021.
Claims priority of application No. 2019-137318 (JP), filed on Jul. 25, 2019.
Prior Publication US 2022/0358749 A1, Nov. 10, 2022
Int. Cl. G06K 9/00 (2022.01); G06T 7/00 (2017.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01); G06V 20/56 (2022.01)
CPC G06V 10/774 (2022.01) [G06T 7/0002 (2013.01); G06V 10/82 (2022.01); G06V 20/56 (2022.01)] 20 Claims
OG exemplary drawing
 
1. An inference apparatus, comprising:
a processor configured with a program to perform operations comprising:
operation as a first data obtainer configured to obtain predetermined target data usable in predetermined inference;
operation as a second data obtainer configured to obtain environment data about a target environment in which the predetermined inference is to be performed;
operation as a result obtainer configured to provide the target data to a plurality of inference models and obtain an inference result for the target data from each of the plurality of inference models, each of the plurality of inference models being derived from local learning data obtained in a different environment, each of the plurality of inference models being configured to perform the predetermined inference using the target data; and
operation as a combiner configured to combine the inference result from each of the plurality of inference models under a combining rule to generate an inference result for the target data in the target environment, wherein
the combining rule comprises a plurality of combining parameters, each of the plurality of combining parameters defining a priority on an inference result from a corresponding inference model of the plurality of inference models in the target environment,
combining the inference result from each of the plurality of inference models under the combining rule including comprises
determining a value of each of the plurality of combining parameters using the obtained environment data,
weighting the inference result from each of the plurality of inference models using the determined value of each of the plurality of combining parameters, and
combining the weighted inference result from each of the plurality of inference models; and
the processor is configured with the program to perform operations further comprising operation as an output unit configured to output information about the inference result generated from the combining.