pip3 install pandas pip3 install tqdm pip3 install sklearn pip3 install scikit-image pip3 install numpy pip3 install
pip3 install nose
python3 extract_dominant_color.py video_path azp_path
video_path: the path to the videofile
azp_path: the path to a azp-file, a list of .azp-files or the path to a directory cotaining .azp-files
output_path: the path for the output .txt-file that should contain the dominant colors, has to include the filename as a .txt-file,default = dominant_colors.txt
resolution_width: set the resolution width of the videofile, the resolution scales automatically to 16:9,default = 200
bin_threshold: set the percentage (0-100) a color has to reach to be returned,default = 5
colors_to_return: set how many colors should be returned at maximum,default = 5
colors_txt: path to a .txt-file containing colors, the file must be in the format 'black:(0,0,0) new line red:(255,0,0) etc',default are a list of 40 colors hardcoded
what_to_process: decide if the dominant colors should be processed per segment or a whole scene, default is segment, switch to scene with 'scene'
target_colorspace: change the colorspace of the video, for now only supports HSV and cie-lab