classification of surgical video scene
Classify surgery scene into 11 cases in real time.
├── LICENSE
├── README.md <- The top-level README for developers using this project.
├── data
│ ├── external <- Data from third party sources.
│ ├── interim <- Intermediate data that has been transformed.
│ ├── processed <- The final, canonical data sets for modeling.
│ └── raw <- The original, immutable data dump.
│
│
├── models <- Trained and serialized models(.h5py), model predictions, or model summaries
│
├── notebooks <- Jupyter notebooks. Naming convention is a number (for ordering),
│ the creator's initials, and a short `-` delimited description, e.g.
│ `1.0-jqp-initial-data-exploration`.
│
│
├── reports <- Generated analysis as HTML, PDF, LaTeX, etc.
│ ├── figures <- Generated graphics and figures to be used in reporting
│ └── log <- accuracy and loss history
│
├── requirements.txt <- The requirements file for reproducing the analysis environment, e.g.
│ generated with `pip freeze > requirements.txt`
│
└── src <- Source code for use in this project.
├── __init__.py <- Makes src a Python module
│
├── reader <- Scripts to download or generate data
│ └── make_dataset.py
│
├── generators <- Scripts to turn raw data into features for modeling
│ └── build_features.py
│
├── models <- Scripts to train models and then use trained models to make
│ │ predictions
│ └── cnn_lstm.py
│
├── visualization <- Scripts to create exploratory and results oriented visualizations
│ └── visualize.py
│
└── train.py
Surgery Movie. Its duration is about two hours.
All data used in this project is in the DLB1 extended HDD (12T).
/mnt
└──hdd
└─data
├── movies
│ ├── case05_HEIGHT720_FPS30.mp4
│ ├── case06_HEIGHT720_FPS30.mp4
│
└── movie_frames
├── case05_HEIGHT720_FPS30
│ ├─ 1.jpg
│ ├─ 2.jpg
│
├── case06_HEIGHT720_FPS30
│ ├─ 1.jpg
│ ├─ 2.jpg
│
- Classify each frame(not time series) using Xception or Inceptionv3, ResNet.
- 3D CNN (time series)
Hiroki Matsuzaki