Get video captioning and use them to analyze the plot, theme and genre. Finally, merge videos that contain the same plot, theme, and genre into a longer video.
goal: use the source video(a lot of full film and video) to generate a personal preference promotional video
user select genre, theme, plot of videos and system generate the short personal preference promotional video. pytorch implementation of video captioning.
- frontend: django+HTML+javascript, realize the voice control(annyang) and frontend maschine human interaction
- backend: django+mysql, to manage data(such as video path, caption, video inforamtion)
- video_caption_generate: pytorch
- video classification(humar behavior): pytorch
- Multi-label universal video classification
- Spracherkennung: frontend+backend
- video_caption: video caption, video classification(humar behavior), Multi-label universal video classification
- master: frontend html file+full version code
In this Branch need to realize the human maschine interaction, use the django framework and voice control(annyang javescript)
- interface design @YueNing @kevin
- backend data management and data structur design @kevin @brankyy Not yet implemented
- voice control @brankyy @YueNing Not yet implemented
In this Branch need to realize the video process code,
- video caption(Video-based semantic content) @YueNing
- video classification(Human behavior recognition) @YueNing
- Multi-label universal video classification(Video-based semantic content) @YueNing @brankyy Not yet implemented
In this Branch merge all project code to here to get the final project code