US 11,861,842 B2
Information processing method, program, and information processing system
Hirotaka Ishikawa, Kanagawa (JP); Kanji Ogawa, Kanagawa (JP); Toshiyuki Yamauchi, Tokyo (JP); Akihiro Muto, Kanagawa (JP); and Tomoki Oooka, Kanagawa (JP)
Assigned to Sony Semiconductor Solutions Corporation, Kanagawa (JP)
Appl. No. 17/267,963
Filed by Sony Semiconductor Solutions Corporation, Kanagawa (JP)
PCT Filed Sep. 9, 2019, PCT No. PCT/JP2019/035395
§ 371(c)(1), (2) Date Feb. 11, 2021,
PCT Pub. No. WO2020/059564, PCT Pub. Date Mar. 26, 2020.
Claims priority of application No. 2018-174000 (JP), filed on Sep. 18, 2018.
Prior Publication US 2021/0264608 A1, Aug. 26, 2021
Int. Cl. G06T 7/00 (2017.01); G06T 7/11 (2017.01); G06N 3/08 (2023.01); G06V 10/25 (2022.01); G06V 20/58 (2022.01)
CPC G06T 7/11 (2017.01) [G06N 3/08 (2013.01); G06V 10/25 (2022.01); G06T 2207/10024 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20132 (2013.01); G06V 20/58 (2022.01)] 20 Claims
OG exemplary drawing
 
1. An information processing method comprising:
automatically cropping a region including an object, from a material image to generate an automatically cropped image; and
performing learning related to detection of the object on the basis of the automatically cropped image, wherein
generating the automatically cropped image uses an automatic cropper generated by learning based on manually cropped images obtained by manually cropping from a set of learning material images,
the automatic cropper is generated on the basis of learned parameters obtained as a result of learning by a neural network to which the manually cropped images are input, and
generating the automatically cropped image includes identifying an interim region including the object in the material image and automatically cropping a detailed region including the object from the interim region.