Robots' Vision Humanization Through Machine-Learning Based Artificial Visual Attention

If the main challenge of robotics during the industrial air of 19th century has consisted of automating repetitive tasks and the sophistication of these machines through digitization of these robots throughout the 20th century, the challenge of robotics in the current century will be to make cohabit humans and robots in the same living space. However, robots would not succeed in seamlessly integrate the humans' universe without developing the ability of perceiving similarly to humans the environment that they are supposed to share with them. In such a context, fitting the skills of the natural vision is an appealing perspective for autonomous robotics dealing with and prospecting Human-Robot interaction. The main goal of the present article is to debate the plausibility and the reality of humanizing the robots behavior focusing the perception of the surrounding environment. An implementation of the developed concept on a real humanoid robot nourishes the presented results and the related discussions.