Video classification using Dataset: HockeyFights
c3d-pytorch
├── extract_frames.py
├── hockey_data.py
├── hockey_model.py
├── test.txt
├── train.txt
├── log.py
├── hockey_train.h5
├── hockey_test.h5
├── data
│ ├── fi100_xvid
│ ├── fi101_xvid
│ ├── fi102_xvid
│ │ ├── 0.jpg
│ │ ├── 1.jpg
│ │ ├── 2.jpg
│ │ ├── ...
│ ├── ...
├── video
│ ├── HockeyFights
│ │ ├── fi100_xvid.avi
│ │ ├── fi101_xvid.avi
│ │ ├── fi102_xvid.avi
│ │ ├── ...
Please refer to data_preparation.md for a general knowledge of data preparation. The supported datasets are listed in supported_datasets.md.
We also share our Kinetics-400 annotation file k400_val, k400_train for better comparison.
To train a video recognition model with pre-trained image models (for Kinetics-400 and Kineticc-600 datasets), run:
Colaboratory file: C3D_Hockey
Model | Input size | acc |
---|---|---|
C3D | 16 x 70 x 110 | 93.50 |
R21D | 8 x 112 x 112 | 96.00 |
P3D | 16 x 160 x 160 | 98.00 |
ARTNet | 16x112x112 | 98.00 |