c3d model link1: https://drive.google.com/file/d/19NWziHWh1LgCcHU34geoKwYezAogv9fX/view?usp=sharing
c3d model link2: https://mega.nz/#!diIBFITA!Wr2LaRURPp8GLVeDjJ9jgJcKIZ48rhJqEjYtbIaEXic
Put your own videos in "dataset_videos" folder in the following hiearchy:
dataset_videos
├── Dragging
│ ├── dragging_1.avi
│ └── ...
├── Kindnapping
│ ├── kidnapping_1.avi
│ └── ...
└── Normal
│ ├── normal_1.avi
│ └── ...
Use image_frm_video.py to extract frames from video put them in the "VAR/dataset_images" folder in following order:
VAR/dataset_images
├── Dragging
│ ├── dragging_1
│ │ ├── 00001.jpg
│ │ └── ...
│ └── ...
├── Kindnapping
│ ├── Kindnapping
│ │ ├── 00001.jpg
│ │ └── ...
│ └── ...
└── Normal
│ ├── Normal
│ │ ├── 00001.jpg
│ │ └── ...
│ └── ...
run "python train.py" to train data
run "python inference.py" to test the model data on any video
Results of our work is here in YouTube video:https://www.youtube.com/watch?v=i4EloWzGLAA
run the code and enjoy