MIT researchers have used probabilistic programming and inverse graphics to develop a framework that enables computer vision systems to more accurately explain 3D scenes observed in 2D images.
Computer vision systems sometimes make inferences about a scene that fly in the face of common sense. For example, if a robot were processing a scene of a
Researchers at MIT and elsewhere have demonstrated a novel technique that vastly increases the speed of rendering 3D scenes from images by using a neural network to reconstruct the 360-degree light field of the underlying 3D scene.