The era of big data and machine learning has rapidly produced techniques to process and extract patterns from large amounts of data. Videos are the epitome of big data by being inherently high dimensional both spatially and temporally. In Neuroscience, murine experiments often include a camera capturing the behavior of head-fixed mice and a camera capturing the activity of the brain while performing the task. Recent work has shown the high explanatory power of linear models using the mouse's facial movements to predict brain activity while preforming a task. The goal of this work is to show the performance of linear models and recurrent deep learning models on predicting brain activity at different frame rates and using deep learning to extract lower dimensional representation of the behavioral states of the mouse from the video for more efficient processing.