A major challenge in clinical In-Vitro Fertilization (IVF) is selecting the\nhighest quality embryo to transfer to the patient in the hopes of achieving a\npregnancy. Time-lapse microscopy provides clinicians with a wealth of\ninformation for selecting embryos. However, the resulting movies of embryos are\ncurrently analyzed manually, which is time consuming and subjective. Here, we\nautomate feature extraction of time-lapse microscopy of human embryos with a\nmachine-learning pipeline of five convolutional neural networks (CNNs). Our\npipeline consists of (1) semantic segmentation of the regions of the embryo,\n(2) regression predictions of fragment severity, (3) classification of the\ndevelopmental stage, and object instance segmentation of (4) cells and (5)\npronuclei. Our approach greatly speeds up the measurement of quantitative,\nbiologically relevant features that may aid in embryo selection.\n