HumanGen: Generating Human Radiance Fields with Explicit Priors

TL;DR

HumanGen introduces a novel 3D human radiance field generation method that combines explicit priors, hybrid feature representation, and disentangled geometry and appearance synthesis to achieve high-quality, view-consistent, and editable human avatars.

Contribution

It presents a new scheme integrating 2D and 3D priors via an anchor image, enabling detailed, realistic, and editable 3D human radiance fields with state-of-the-art quality.

Findings

Achieves detailed geometry and high-quality textures.

Demonstrates superior free-view rendering performance.

Enables seamless integration of 2D latent editing into 3D generation.

Abstract

Recent years have witnessed the tremendous progress of 3D GANs for generating view-consistent radiance fields with photo-realism. Yet, high-quality generation of human radiance fields remains challenging, partially due to the limited human-related priors adopted in existing methods. We present HumanGen, a novel 3D human generation scheme with detailed geometry and $\text{360}^{\circ}$ realistic free-view rendering. It explicitly marries the 3D human generation with various priors from the 2D generator and 3D reconstructor of humans through the design of "anchor image". We introduce a hybrid feature representation using the anchor image to bridge the latent space of HumanGen with the existing 2D generator. We then adopt a pronged design to disentangle the generation of geometry and appearance. With the aid of the anchor image, we adapt a 3D reconstructor for fine-grained details synthesis and propose a two-stage blending scheme to boost appearance generation. Extensive experiments demonstrate our effectiveness for state-of-the-art 3D human generation regarding geometry details, texture quality, and free-view performance. Notably, HumanGen can also incorporate various off-the-shelf 2D latent editing methods, seamlessly lifting them into 3D.