This robot from the Learning Algorithms and Systems Laboratory draws from a photo.
The vision system detects the presence of a person and snaps a photo. The robot then asks for it’s quill pen. The robot then starts outlining the contours first, then filling in the details later.
Finally the robot draws a frame around the portrait, signs it and tells you it is done. The robot has 3 drawing modes one can choose from. Two modes are line based and the third consists of dots.