Generative Deformable Radiance Fields for Disentangled Image Synthesis of Topology-Varying Objects

TL;DR

This paper introduces a generative model for topology-varying objects that produces deformable radiance fields with disentangled shape and appearance, enabling high-quality 3D object synthesis and editing from monocular images.

Contribution

It proposes a novel deformable radiance field model that disentangles shape and appearance without supervision, facilitating realistic 3D object generation and manipulation from monocular images.

Findings

Successfully disentangles shape and appearance in topology-varying objects

Achieves high-quality 3D reconstruction and editing from single images

Works effectively on both synthetic and real monocular images

Abstract

3D-aware generative models have demonstrated their superb performance to generate 3D neural radiance fields (NeRF) from a collection of monocular 2D images even for topology-varying object categories. However, these methods still lack the capability to separately control the shape and appearance of the objects in the generated radiance fields. In this paper, we propose a generative model for synthesizing radiance fields of topology-varying objects with disentangled shape and appearance variations. Our method generates deformable radiance fields, which builds the dense correspondence between the density fields of the objects and encodes their appearances in a shared template field. Our disentanglement is achieved in an unsupervised manner without introducing extra labels to previous 3D-aware GAN training. We also develop an effective image inversion scheme for reconstructing the radiance field of an object in a real monocular image and manipulating its shape and appearance. Experiments show that our method can successfully learn the generative model from unstructured monocular images and well disentangle the shape and appearance for objects (e.g., chairs) with large topological variance. The model trained on synthetic data can faithfully reconstruct the real object in a given single image and achieve high-quality texture and shape editing results.