Title:
VIDEO GENERATION METHOD AND APPARATUS AND COMPUTER-READABLE STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2024/056078
Kind Code:
A1
Abstract:
The embodiments of the present application disclose a video generation method and apparatus and a computer-readable storage medium. The method comprises: acquiring initial state data and audio data of a virtual subject; extracting audio features from the audio data, and encoding the audio data to obtain encoded features; fusing the audio features and the encoded features to obtain fused audio features of the virtual subject; acquiring updated state data of the virtual subject according to the fused audio features and the initial state data; and generating video data according to the updated state data and the audio data. Thus, matching accuracy of a target part in the video data and the audio data is improved, thereby enhancing the video generation accuracy of the virtual subject, and further increasing the video generation efficiency of the virtual subject.
Inventors:
SU GUINAN (CN)
WU GAO (CN)
LI ZHIFENG (CN)
LIU WEI (CN)
CUI YUSHI (CN)
WU GAO (CN)
LI ZHIFENG (CN)
LIU WEI (CN)
CUI YUSHI (CN)
Application Number:
PCT/CN2023/119140
Publication Date:
March 21, 2024
Filing Date:
September 15, 2023
Export Citation:
Assignee:
TENCENT TECH SHENZHEN CO LTD (CN)
International Classes:
G10L21/10; G06T13/40; G06T15/00; G10L25/03; H04N21/439
Domestic Patent References:
| WO2021073416A1 | 2021-04-22 |
Foreign References:
| CN113299312A | 2021-08-24 | |||
| CN110459241A | 2019-11-15 | |||
| CN113313797A | 2021-08-27 | |||
| CN113901894A | 2022-01-07 | |||
| CN202211127727A | 2022-09-16 |
Attorney, Agent or Firm:
SHENPAT INTELLECTUAL PROPERTY AGENCY (CN)
Download PDF: