Self-driving cars can be defined as vehicles that do not require human intervention for navigational purposes. Perception systems allow the vehicle to be aware of its surroundings: obtaining information, evaluating it and taking the optimal action at each time step. Fully Autonomous Vehicles obtain all the information of the surroundings from its sensors.
This means that it is necessary to design systems that are robust on the evaluation of information, producing a consistent result over time. The state-of-the-art for semantic segmentation evaluates the input image on a frame-by-frame basis. The goal of this assignment is to study the extension of semantic image segmentation applications with methods that include information of previous time states into the present evaluation.