Multi-Plane Program Induction with 3D Box Priors
This work addresses the problem of inferring a holistic, structured 3D-aware scene representation from a single image for interactive image editing, which is significant for computer vision and graphics researchers.
This paper introduces Box Program Induction (BPI), a method that infers a program-like scene representation from a single image. It simultaneously models repeated structures on multiple 2D planes, their 3D position and orientation, and camera parameters, assuming a 3D box prior (inner or outer view).
We consider two important aspects in understanding and editing images: modeling regular, program-like texture or patterns in 2D planes, and 3D posing of these planes in the scene. Unlike prior work on image-based program synthesis, which assumes the image contains a single visible 2D plane, we present Box Program Induction (BPI), which infers a program-like scene representation that simultaneously models repeated structure on multiple 2D planes, the 3D position and orientation of the planes, and camera parameters, all from a single image. Our model assumes a box prior, i.e., that the image captures either an inner view or an outer view of a box in 3D. It uses neural networks to infer visual cues such as vanishing points, wireframe lines to guide a search-based algorithm to find the program that best explains the image. Such a holistic, structured scene representation enables 3D-aware interactive image editing operations such as inpainting missing pixels, changing camera parameters, and extrapolate the image contents.