Neural Human Performer: Learning Generalizable Radiance Fields for Human Performance Rendering

Abstract

In this paper, we aim at synthesizing a free-viewpoint video of an arbitrary human performance using sparse multi-view cameras. Recently, several works have addressed this problem by learning person-specific neural radiance fields (NeRF) to capture the appearance of a particular human. In parallel, some work proposed to use pixel-aligned features to generalize radiance fields to arbitrary new scenes and objects. Adopting such generalization approaches to humans, however, is highly challenging due to the heavy occlusions and dynamic articulations of body parts. To tackle this, we propose Neural Human Performer, a novel approach that learns generalizable neural radiance fields based on a parametric human body model for robust performance capture. Specifically, we first introduce a temporal transformer that aggregates tracked visual features based on the skeletal body motion over time. Moreover, a multi-view transformer is proposed to perform cross-attention between the temporally-fused features and the pixel-aligned features at each time step to integrate observations on the fly from multiple views. Experiments on the ZJU-MoCap and AIST datasets show that our method significantly outperforms recent generalizable NeRF methods on unseen identities and poses. The video results and code are available at https://youngjoongunc.github.io/nhp.

References in corpus (10)

• Stereo Magnification: Learning View Synthesis using Multiplane Images
• Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations
• pixelNeRF: Neural Radiance Fields from One or Few Images
• Portrait Neural Radiance Fields from a Single Image
• IBRNet: Learning Multi-View Image-Based Rendering
• PVA: Pixel-aligned Volumetric Avatars
• Space-time Neural Irradiance Fields for Free-Viewpoint Video
• Convolutional Occupancy Networks
• Neural Sparse Voxel Fields
• Reconstructing 3D Human Pose by Watching Humans in the Mirror

Cited by in corpus (4)

• Adaptive Multi-view and Temporal Fusing Transformer for 3D Human Pose Estimation
• Physics Informed Neural Fields for Smoke Reconstruction with Sparse Data
• Artemis: Articulated Neural Pets with Appearance and Motion synthesis
• PACE: Data-Driven Virtual Agent Interaction in Dense and Cluttered Environments