Video highlight extraction method and system and storage medium

Abstract

The invention relates to a video highlight extraction method and system and a storage medium. The method comprises the following steps: acquiring a to-be-processed network class video and teacher-student interaction characteristics, and dividing the to-be-processed network class video into a plurality of target videos, respectively performing facial expression analysis, eye movement attention analysis and gesture and body movement analysis on the pictures corresponding to all the frames of the target video to obtain a visual feature set of students and a visual feature set of teachers in the pictures corresponding to all the frames; determining the timeliness of student feedback according to the time interval between the voice segment corresponding to the student and the voice segment corresponding to the teacher in the audio of the target video; performing voice recognition and keyword extraction on the voice segments corresponding to the students and the teachers to determine fluency of teacher languages, fluency of student languages and correctness of teaching knowledge; and according to the priority of each target video, determining a highlight in the to-be-processed network class video. The method and system have high interpretability, and a user can conveniently adjust parameters according to requirements.

Description

technical field [0001] The invention relates to the field of video analysis and clipping, in particular to a method, system and storage medium for extracting highlights of a video. Background technique [0002] With the development of remote classroom technology, students, teachers, and course sales staff hope to evaluate and promote the exciting content in the classroom through short videos. Manual editing, this method is time-consuming, and the quality of the edited content is affected by unclear editing standards. Human factors have a greater impact, and educational institutions also hope to use automated methods to produce course highlights to promote their courses to reduce the cost of manual editing. Existing automatic video editing technologies include: end-to-end video editing through neural networks , this method is suitable for videos with fast lens changes, such as movies, etc. For online course videos that emphasize course content, these end-to-end editing models...

AI Technical Summary

Problems solved by technology

[0002] With the development of remote classroom technology, students, teachers, and course sales staff hope to evaluate and promote the exciting content in the classroom through short videos. Manual editing, this method is time-consuming, and the quality of the edited content is affected by unclear editing standards. Human factors have a greater impact, and educational institutions also hope to use automated methods to produce course highlights to promote their courses to reduce the cost of manual editing. Existing automatic video editing technologies include: end-to-end video editing through neural networks , this method is suitable for videos with fast lens changes, such as movies, etc. For online course videos that emphasize course content, these end-to-end editing models are not very interpretable, and need to be retrained when the demand for videos changes Model, heavy workload and high cost

Images