PixelHuman: Animatable Neural Radiance Fields from Few Images

TL;DR

PixelHuman introduces a new human rendering model that can generate animatable 3D human scenes from only a few images, generalizing to unseen identities, views, and poses with state-of-the-art results.

Contribution

It is the first method to produce animatable human scenes from limited images without scene-specific training, using a novel combination of canonical representation and pose-aware features.

Findings

Achieves state-of-the-art multiview synthesis from few images.

Successfully generalizes to unseen identities and poses.

Operates efficiently without per-scene training.

Abstract

In this paper, we propose PixelHuman, a novel human rendering model that generates animatable human scenes from a few images of a person with unseen identity, views, and poses. Previous work have demonstrated reasonable performance in novel view and pose synthesis, but they rely on a large number of images to train and are trained per scene from videos, which requires significant amount of time to produce animatable scenes from unseen human images. Our method differs from existing methods in that it can generalize to any input image for animatable human synthesis. Given a random pose sequence, our method synthesizes each target scene using a neural radiance field that is conditioned on a canonical representation and pose-aware pixel-aligned features, both of which can be obtained through deformation fields learned in a data-driven manner. Our experiments show that our method achieves state-of-the-art performance in multiview and novel pose synthesis from few-shot images.