US 12,306,867 B2
Production method of multimedia work, apparatus, and computer-readable storage medium
Xiaojuan Cai, Beijing (CN); Xuchen Song, Los Angeles, CA (US); Gen Li, Beijing (CN); Haoyuan Zhong, Los Angeles, CA (US); Weishu Mo, Beijing (CN); and Hui Li, Beijing (CN)
Assigned to LEMON INC., Grand Cayman (KY)
Filed by Lemon Inc., Grand Cayman (KY)
Filed on Dec. 20, 2022, as Appl. No. 18/069,031.
Application 18/069,031 is a continuation of application No. PCT/SG2021/050470, filed on Aug. 11, 2021.
Claims priority of application No. 202010901767.7 (CN), filed on Aug. 31, 2020.
Prior Publication US 2023/0131850 A1, Apr. 27, 2023
Int. Cl. G06F 16/00 (2019.01); G06F 16/483 (2019.01); G06F 16/683 (2019.01); G06N 3/02 (2006.01)
CPC G06F 16/483 (2019.01) [G06F 16/683 (2019.01); G06N 3/02 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A production method of multimedia work, comprising:
acquiring a target audio and at least one piece of multimedia information, wherein the at least one piece of multimedia information comprises at least one selected from the group consisting of a picture or a video;
determining a matching degree between the target audio and each of the at least one piece of multimedia information to obtain at least one matching degree, sorting the at least one piece of multimedia information in an order of the at least one matching degree from high to low, and taking a first preset number of pieces of multimedia information ranking in top order as target multimedia information;
determining image quality of each picture in a target multimedia information, sorting each picture in the target multimedia information in an order of the image quality of each picture from high to low, and taking a second preset number of pictures ranking in top order as target pictures;
synthesizing a multimedia work according to the target pictures and the target audio,
wherein the determining image quality of each picture in the target multimedia information, comprises:
inputting the target pictures into an image quality prediction model that is pre-trained, to obtain image quality of the target pictures output by the image quality prediction model.