1. Make sure Python >= 3.7 and Git are pre-installed.

2. Clone this repository.

$ git clone [email protected]:liberaliscomputing/video-object-detection.git
$ cd video-object-detection

3. Create and activate a virtual environment. Install dependencies.

$ python3 -m venv venv && source venv/bin/activate
(venv) $ pip install --upgrade pip && pip install -r requirements.txt

4. Download the PointRend model.

(venv) $ wget --quiet https://github.com/ayoolaolafenwa/PixelLib/releases/download/0.2.0/pointrend_resnet50.pkl

5. This script requires two arguments.

(venv) $ python detect.py -h

• SOURCE: A path to source video
• OUTPUT: A path to output video

The following optional arguments can be passed in as well:

• --model: A model with pre-trained weights, defaults to "pointrend_resnet50.pkl"
• --frames_per_second: A Parameter that sets the number of frames per second for the output video, defaults to 3

6. Run the following command with arguments.

(venv) $ python detect.py source.mov output.mov [--frame_per_second 10]

Source

Output