Encoding Longer-Term Contextual Information with Predictive Coding and Ego-Motion

Figure 2

Data Collected from VRep Simulation.