Thesis 2021: VIDEO SUMMARIZING AT THE HAPPIEST MOMENT VIA FACE BEAUTIFICATION

1. Download conda

See this instruction to download conda.

2. Set up environment

• Download source code:

git clone https://github.com/tuandq-cs/thesis-2021.git

• Install dependences and create a new conda environment:

conda env create -f environment.yml

3. Get pretrained models

• Set current working directory to root:

cd thesis-2021/

• Install pretrained model
• Extract file:

unzip models.zip

4. Run

• Activate the environment:

conda activate $YOUR_ENV_NAME

• Finally, put your video path into this script:

python main.py --video_path=$YOUR_VIDEO_PATH

• Result will be store in inference folder inferences/
• Folders in inference folder are formated by: year_month_day_hour_min
• The lastest folder is our result.

Alternatives

Or make it easier by looking at this Google Colab file