Can NeRFs See without Cameras?

TL;DR

This paper demonstrates that Neural Radiance Fields (NeRFs) can be adapted to learn from multipath radio frequency signals, enabling indoor environment inference like floorplan reconstruction from WiFi data.

Contribution

The authors introduce a redesign of NeRFs to learn from multipath RF signals, allowing environment inference without traditional camera data.

Findings

NeRFs can be trained on multipath RF signals to infer indoor environments.

The method produces promising floorplan reconstructions from sparse WiFi measurements.

Applications include indoor signal prediction and basic ray tracing.

Abstract

Neural Radiance Fields (NeRFs) have been remarkably successful at synthesizing novel views of 3D scenes by optimizing a volumetric scene function. This scene function models how optical rays bring color information from a 3D object to the camera pixels. Radio frequency (RF) or audio signals can also be viewed as a vehicle for delivering information about the environment to a sensor. However, unlike camera pixels, an RF/audio sensor receives a mixture of signals that contain many environmental reflections (also called "multipath"). Is it still possible to infer the environment using such multipath signals? We show that with redesign, NeRFs can be taught to learn from multipath signals, and thereby "see" the environment. As a grounding application, we aim to infer the indoor floorplan of a home from sparse WiFi measurements made at multiple locations inside the home. Although a difficult inverse problem, our implicitly learnt floorplans look promising, and enables forward applications, such as indoor signal prediction and basic ray tracing.