US 12,136,258 B2
Information processing apparatus, method for operating information processing apparatus, and operating program of information processing apparatus
Yasushi Shiraishi, Kanagawa (JP)
Assigned to FUJIFILM CORPORATION, Tokyo (JP)
Filed by FUJIFILM CORPORATION, Tokyo (JP)
Filed on Feb. 28, 2022, as Appl. No. 17/683,259.
Application 17/683,259 is a continuation of application No. PCT/JP2020/017405, filed on Apr. 22, 2020.
Claims priority of application No. 2019-177219 (JP), filed on Sep. 27, 2019.
Prior Publication US 2022/0189149 A1, Jun. 16, 2022
Int. Cl. G06K 9/00 (2022.01); G06V 10/10 (2022.01); G06V 10/26 (2022.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 20/69 (2022.01)
CPC G06V 10/776 (2022.01) [G06V 10/16 (2022.01); G06V 10/26 (2022.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 20/695 (2022.01); G06V 20/698 (2022.01)] 12 Claims
OG exemplary drawing
 
1. An information processing apparatus, comprising a processor that is configured to:
acquire an output image output from a trained model as a result of causing the trained model to perform semantic segmentation in which discrimination of a class which is a type of an object appearing in an input image is performed on a pixel-by-pixel basis;
evaluate validity of the semantic segmentation based on the output image; and
perform control such that an evaluation result indicating that the semantic segmentation does not have validity is displayed in a case of evaluating that the semantic segmentation does not have validity,
wherein the processor calculates an image feature value from the output image, and evaluates the validity of the semantic segmentation by using the image feature value, the image feature value being associated with a pre-selected image feature that is different from image features employed by the trained model that performs semantic segmentation.