Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
DEEP-LEARNING-BASED VIDEO STREAMING METHOD AND SYSTEM THEREFOR
Document Type and Number:
WIPO Patent Application WO/2022/270659
Kind Code:
A1
Abstract:
A deep-learning-based video streaming method using image captioning, and a system therefor are disclosed. According to an embodiment of the present invention, a deep-learning-based video streaming method using image captioning comprises steps in which: a system splits a video into a plurality of shots; the system determines at least one frame selected from among a plurality of frames included in each of the split shots; at least one shot text for each of the split shots is generated through image captioning corresponding to the selected frame determined by the system; and the video is edited on the basis of the at least one shot text for each of the split shots generated by the system.

More Like This:
WO/2015/116614SYSTEMS AND METHODS FOR SPACIAL AND TEMPORAL EXPERIMENTATION ON CONTENT EFFECTIVENESS
WO/2016/104932IMAGE DISPLAY APPARATUS AND IMAGE DISPLAY METHOD
JP4709778Information processing device and recording control method
Inventors:
SHIM CHOONG SUP (KR)
HA YOUNG KWANG (KR)
Application Number:
PCT/KR2021/008062
Publication Date:
December 29, 2022
Filing Date:
June 28, 2021
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
DEEP HIGH INC (KR)
International Classes:
H04N21/472; G06N3/08; H04N21/845; H04N21/854
Foreign References:
KR20190120106A2019-10-23
KR20190137359A2019-12-11
KR20200001246A2020-01-06
KR20200023013A2020-03-04
KR20210064587A2021-06-03
Attorney, Agent or Firm:
THE WAVE IP LAW FIRM (KR)
Download PDF:
View/Download PDF      PDF Help



 
Previous Patent: SMART HELMET

Next Patent: VIDEO STREAM PROCESSING METHOD BASED ON DEEP LEARNING-BASED CENTRAL OBJECT, AND SYSTEM THEREFOR