US 12,277,777 B2
Information processing device and information processing method
Hsingying Ho, Tokyo (JP); Christopher Wright, Lausanne (CH); Nicholas Walker, Lausanne (CH); and Bernadette Elliot-Bowman, Lausanne (CH)
Assigned to SONY GROUP CORPORATION, Tokyo (JP)
Appl. No. 17/998,179
Filed by SONY GROUP CORPORATION, Tokyo (JP)
PCT Filed Apr. 7, 2021, PCT No. PCT/JP2021/014780
§ 371(c)(1), (2) Date Nov. 8, 2022,
PCT Pub. No. WO2021/235126, PCT Pub. Date Nov. 25, 2021.
Claims priority of application No. 2020-087122 (JP), filed on May 19, 2020.
Prior Publication US 2023/0298357 A1, Sep. 21, 2023
Int. Cl. G06V 20/58 (2022.01); G06T 7/70 (2017.01); G06T 11/00 (2006.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/17 (2022.01); H04R 1/08 (2006.01); H04R 3/00 (2006.01); H04R 23/00 (2006.01)
CPC G06V 20/58 (2022.01) [G06T 7/70 (2017.01); G06T 11/00 (2013.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/17 (2022.01); H04R 1/08 (2013.01); H04R 3/00 (2013.01); H04R 23/008 (2013.01); G06T 2207/10032 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30252 (2013.01)] 16 Claims
OG exemplary drawing
 
1. An information processing device, comprising:
a central processing unit (CPU) configured to:
estimate, based on an input image, a class of an object that is present in a real environment, wherein
the object includes an acoustically useful object that has an acoustic feature, and
the real environment corresponds to a range of the input image;
collect acoustic data of the acoustically useful object;
estimate a class of the acoustically useful object based on the collected acoustic data;
generate a first estimator based on machine learning in which the collected acoustic data is input at a specific time and a first image related to the acoustically useful object is output;
obtain a second image of the acoustic useful object, wherein the second image is captured at the specific time;
reduce a difference between the first image and the second image based on the generated first estimator; and
create a composite image based on the reduced difference, wherein the composite image displays the acoustically useful object.