HyperFields: Towards Zero-Shot Generation of NeRFs from Text

TL;DR

HyperFields is a novel method that enables zero-shot and few-shot generation of diverse NeRF scenes from text prompts using a dynamic hypernetwork, significantly speeding up scene synthesis.

Contribution

We propose HyperFields, a dynamic hypernetwork approach that learns a general text-to-NeRF mapping, allowing rapid zero-shot and few-shot scene generation from text.

Findings

Enables single-pass generation of over 100 scenes from text.

Capable of zero-shot scene synthesis, including out-of-distribution scenes.

Faster scene synthesis, 5 to 10 times quicker than existing methods.

Abstract

We introduce HyperFields, a method for generating text-conditioned Neural Radiance Fields (NeRFs) with a single forward pass and (optionally) some fine-tuning. Key to our approach are: (i) a dynamic hypernetwork, which learns a smooth mapping from text token embeddings to the space of NeRFs; (ii) NeRF distillation training, which distills scenes encoded in individual NeRFs into one dynamic hypernetwork. These techniques enable a single network to fit over a hundred unique scenes. We further demonstrate that HyperFields learns a more general map between text and NeRFs, and consequently is capable of predicting novel in-distribution and out-of-distribution scenes -- either zero-shot or with a few finetuning steps. Finetuning HyperFields benefits from accelerated convergence thanks to the learned general map, and is capable of synthesizing novel scenes 5 to 10 times faster than existing neural optimization-based methods. Our ablation experiments show that both the dynamic architecture and NeRF distillation are critical to the expressivity of HyperFields.