Neural Point Catacaustics for Novel-View Synthesis of Reflections
DescriptionView-dependent effects such as reflections pose a substantial challenge for image-based and neural rendering algorithms. Above all, curved reflectors are particularly hard, as they lead to highly non-linear reflection flows as the camera moves. We introduce Neural Point Catacaustics, a new point-based representation to render novel views of scenes with curved reflectors from a set of casually-captured input photos. At the core of our method is a neural warp field that models catacaustic trajectories of reflections, so complex specular effects can be rendered using efficient point splatting in conjunction with a neural renderer. One of our key contributions is the direct representation of reflections with a reflection point cloud which is displaced by the neural warp field, and a primary point cloud optimized in a preprocess, to represent the rest of the scene. Our approach allows interactive high-quality renderings of novel views with accurate reflection flow. Additionally, the direct representation of this reflection flow for the first time enables several forms of scene manipulation in captured scenes, such as reflection editing, cloning of specular objects, reflection tracking across views, and comfortable stereo viewing.
Event Type
Technical Communications
Technical Papers
TimeWednesday, 7 December 202211:00am - 12:30pm KST
Registration Categories