😃 Multi-video-semantic-intelligent-analysis-platform 😋
The Python-based Multi-video semantic intelligent analysis platform integrates various research results of the laboratory in the field of Multi-video understanding at this stage, which is helpful for recording and displaying experimental results and promoting academic exchange activities. It is one of the huge challenges of the project to concentrate the major model frameworks in one system. The design of the platform interface integrates popular elements such as barrage, search, and web page jumps. The entire system integrates rigorous academic research technology and an image display interface. , which is easy for users to operate and understand. The semantic information research of multi-videos can accurately obtain some information of multi-videos such as behaviors, scenes and other information, and the perform have other functions such as popularity prediction.These contents can guide commercial positioning, market pricing, etc., and promote the further development of AI technology.
- Scene estimation
- Event detection
- emotion analysis
- Fashion analysis
- Memory analysis
- Popularity prediction
📺 Presentation of the platform
If it’s helpful for you, please give me a thumbs-up. 👍
- Python 3.5.6
- Tensorflow 1.10
- Pytorch 0.4.0
- ...
- The UI of this platform is written by Pyqt5.
- Regarding the parameters of each network used in the platform, because it exceeds the file size requirement of Github, it is not uploaded. You can train by yourself or contact me. ⬇️
- If you have any questions，you can contact me 📧 curryxin@tju.edu.cn.
- I'm also known to blog
- This is my teammate, you can contact her if you have any questions.
This project was greatly helped by Tianjin University, especially the laboratory teachers and students.
If you are interested in the research direction of our laboratory, you can visit our website Institute of Television and Image Information, Tianjin University