Title:
INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND INFORMATION PROCESSING PROGRAM
Document Type and Number:
WIPO Patent Application WO/2020/080268
Kind Code:
A1
Abstract:
This information processing device (100) comprises: an acquisition unit (132) that uses a trained encoder and obtains a feature value for each piece of partial data constituting first content; a calculation unit (133) that calculates a relative feature value sequence indicating the features of the configuration of the first content, calculating same by calculating relative feature values being feature values that are between partial data, from the feature values for each piece of obtained partial data; and a generation unit (134) that generates second content on the basis of the relative feature value sequence for the first content and a feature value for arbitrary data.
Inventors:
AKAMA TAKETO (JP)
Application Number:
PCT/JP2019/040130
Publication Date:
April 23, 2020
Filing Date:
October 10, 2019
Export Citation:
Assignee:
SONY CORP (JP)
International Classes:
G10H1/00; G06N3/02; G06N20/00
Foreign References:
| JP2005070806A | 2005-03-17 | |||
| JP2018005520A | 2018-01-11 | |||
| JP2011175006A | 2011-09-08 |
Other References:
BRIOT, JEAN-PIERRE ET AL.: "Deep Learning Techniques for Music Generation- A Survey", 5 September 2017 (2017-09-05), pages 1 - 108, XP081423814, Retrieved from the Internet
SHIMIZU, RYO: "Passage, First deep learning programming", FIRST DEEP LEARNING PROGRAMMING, 10 January 2017 (2017-01-10), pages 154 - 157, XP009520939, ISBN: 978-4-7741-8534-7
KOKUBO, YOSHIHITO ET AL.: "Fine-grained Pedestrian Classification by Fusing Multiple CNN Models", IEICE TECHNICAL REPORT, vol. 116, no. 461, 11 February 2017 (2017-02-11), pages 81 - 85, XP009520930
NAGASHIMA, HIDEAKI ET AL.: "Feature Extraction of Painting Style Using Deep Neural Network", PROCEEDINGS OF INFORMATION SCIENCE AND TECHNOLOGY FORUM; FIT 2014, vol. 13, no. 3, 19 August 2014 (2014-08-19), pages 133 - 138, XP009520937
See also references of EP 3716262A4
SHIMIZU, RYO: "Passage, First deep learning programming", FIRST DEEP LEARNING PROGRAMMING, 10 January 2017 (2017-01-10), pages 154 - 157, XP009520939, ISBN: 978-4-7741-8534-7
KOKUBO, YOSHIHITO ET AL.: "Fine-grained Pedestrian Classification by Fusing Multiple CNN Models", IEICE TECHNICAL REPORT, vol. 116, no. 461, 11 February 2017 (2017-02-11), pages 81 - 85, XP009520930
NAGASHIMA, HIDEAKI ET AL.: "Feature Extraction of Painting Style Using Deep Neural Network", PROCEEDINGS OF INFORMATION SCIENCE AND TECHNOLOGY FORUM; FIT 2014, vol. 13, no. 3, 19 August 2014 (2014-08-19), pages 133 - 138, XP009520937
See also references of EP 3716262A4
Attorney, Agent or Firm:
SAKAI INTERNATIONAL PATENT OFFICE (JP)
Download PDF: