The principle of self-supervised machine vision is that you simply load your images and hit ‘Run’ - no parameter tuning needed, no training imagery required. The associated open access publication in Communications Biology is now available: https://rdcu.be/cYMfm Run from start to finish, the code uses consecutive pairs of images to generate unsupervised training data of 'cells' and 'background' via dynamic feature vectors based on optical flow. These self-labeled pixels are then used to generate static feature vectors (e.g. entropy, gradient), which in turn are used to train a classifier model. The training data is updated every image in order to automatically adapt to temporal changes in cell morphologies or background illumination. This work has been accepted for publication in the journal Communications Biology and is currently in press (9/23/22) Included in this upload are 1. Stand-alone graphical user interfaces (GUIs) designed to be used with time-resolved live cell microscopy images (tiffs) for the automated segmentation of cells from background. There is a dedicated stand-alone GUI for Windows, Mac and Linux operating systems (OS). See the GUI_ReadMe.pdf for instructions. 2. The Matlab source code. This code was tested on Matlab v2020a, v2021a and v2022a using commercially available laptop computers running the Windows 10 operating system. The code requires the following additional toolboxes to be installed: Computer Vision Toolbox Image Processing Toolbox Statistics and Machine Learning Toolbox Matlab and the associated toolboxes are available for free 30 day trials if not currently licensed by your institution. See the Matlab_ReadMe.pdf for further instructions.