DPA-Net: Structured 3D Abstraction from Sparse Views via Differentiable Primitive Assembly

arXiv:2404.00875v1 Announce Type: new
Abstract: We present a differentiable rendering framework to learn structured 3D abstractions in the form of primitive assemblies from sparse RGB images capturing a 3D object. By leveraging differentiable volume rendering, our method does not require 3D supervision. Architecturally, our network follows the general pipeline of an image-conditioned neural radiance field (NeRF) exemplified by pixelNeRF for color prediction. As our core contribution, we introduce differential primitive assembly (DPA) into NeRF to output a 3D occupancy field …

3d object abstract abstraction abstractions arxiv assembly cs.cv differentiable form framework general images learn network object pipeline rendering supervision type via