There is a lot of discussion in the news around breakthroughs in Artificial Intelligence. Will Knight’s recent story in WIRED explains an innvotion around using AI algorithms to turn 2D images into amazingly detailed 3D views. The process uses a neural network to create 3D animations using only a few still shot images as a starting point. The algorithm is based on understanding the way light passes through air and then computes “the density and color of points in 3D space.”
This concept, in certain ways, is quite similar to Leonardo Da Vinci’s ability to leverage his careful study of optics and light diffusion to create extraordinary 3D representations in his paintings. Da Vinci had the ability to see an object from a single view and then visualize instantaneously how to mimic that object using the medium of paint as his “color of points” on a 2D surface.
Like the AI algorithm, Da Vinci was emulating physical reality using artificial constructs to convey data in a manner that could be visually interpreted as having clear meaning by others.
Leonardo was also doing an additional operation from a neural processing standpoint. Leonardo was constructing images that were a re-imagined reality that conveyed even richer meaning than the precise data that he was receiving in his eyes. Da Vinci was taking in information and re-interpreting it to convey ideals that could not only be understood as objects in physical space, but that could stimulate intended emotion.
In the midst of the development of these incredible artificial tools, what can we learn from examples of human intelligence to better understand both the similarities and differences of each?