How do we see? What is the immediate object of visual perception, and where is it located? Seventeenth-century thinkers offered contrasted answers to such questions. Kepler's discovery of retinal pictures and Descartes's account of nervous transmission and cerebral picture formation set the stage for early modern discussions of visual perception and sensory ideas. Not only is visual perception an effect of the physical pictures painted in the head, but also, as suggested in Kepler’s famous dictum "ut pictura, ita visio," it seems that the content of our perception is strictly constrained by the content of those pictures. However, to sort out the philosophical signification of these facts was not a straightforward task. Descartes's construal of sensory ideas in the Dioptrics and in the Treatise of man, however influential, was not devoid of ambiguities. Among Descartes's followers, as well as among his critics, it gave rise to a number of various and contradictory readings. Are sensory "ideas" these very material pictures, painted and "sensed" in the space of the brain called the "sensorium," or do we need another account, construing ideas either as intellectual beings, mental items of some sort, or as act of perception directed towards the external object itself and triggered by the motions in the brain? This project showed how this philosophical debate on sensation and ideas (which was instrumental in the making of what will be called later the "mind-body problem") has been shaped and directed by contemporary development in optics, and discussions of visual experience. Far from being a purely metaphysical issue, it has been constantly discussed in relation to a large body of increasing empirical knowledge: new anatomical description of the eye and brain, new descriptions of perceptual experience (perspective vision, instrumented vision, binocular vision, after images, optical illusions) and new experimental discoveries on vision, light, and colors.