We aim at developing robot’s eyes that can sense and recognize the world that we live in. The study includes Computer Vision and Machine Learning, specifically, 3D reconstruction from images, optimization and learning techniques for object and action recognition.