TODO already working for images, disable video controls/info when user loads an image camera already supported, test on OSX append OCR result to text file write installation instructions for python packages in README