Classification of EEG with Recurrent Neural Networks

3-D perception is a task that is growing in popularity in television and entertainment. Algorithms and innovations that mimic 3-D perception are of great importance to those in this industry, and as such they need a metric for how well a particular innovation is working. Electroencephalogram (EEG) recordings are an accurate and objective method of evaluating brain activity, and so the primary task is to use EEG recordings score different methods of mimicking 3-D perception. As a first step in doing this we must find the best features and methods to classify EEG recorded when participants are viewing regular 2D stimuli, and actual 3D stimuli. Hence, in this paper, we explore methods to address the following goal: