- Recognize idol in video using Deep Learning
- YouTube viewer using the result information
- Google Image Search
- Bing Image Search
- Youtube & Face Detection
- refer to face.py
- face detection using dlib
- detected faces are placed in the candidates directory
- candidates are sorted by hand and placed in recognized idol directory
- Remove almost same image and blurred image
- This prevents the number of training data from increasing unnecessarily
- TODO: Efficiency can be improved if there is a tool to sort by image similarity
- Keras Resnet
- MXNet Resnext
- Idol face recognition
- Training
- Prediction
- Visualization
- using t-SNE
- face similarity between idols
- 0.627 on validation set (3000 images)
- 0.9 is desirable and I think it is possible.
- same as food classification settings
- Place text file with key written to secret/google_search_api_key
- API Manager
- https://console.developers.google.com
- You can see the number of queries remaining today
- Limit a day is 100 queries (few...)
- Get cx
- Request URL example
- Download mp4 by running code/collect/youtube_downloader.py
- Run code/video_analyzer.py after assign youtube id
- Run docs\view\list\list_data_maker.py
- Since it is meaningless to learn almost the same face image, we want to eliminate overlapped images by sorting by image similarity
- Make READMEs ogranized
- How to manage the change of group of idol or resignation
- Withdrawal while maintaining a good relationship
- Visible?
- Withdrawal due to causing problems
- Invisible?
- Withdrawal while maintaining a good relationship