PlatoNeRF: 3D Reconstruction in Plato's Cave via Single-View Two-Bounce Lidar

TL;DR

PlatoNeRF introduces a novel single-view 3D reconstruction method using lidar data to model two-bounce light paths, overcoming occlusion and ambiguity issues without relying on priors or controlled lighting.

Contribution

It leverages single-photon lidar data to model two-bounce optical paths with NeRF, enabling accurate 3D reconstruction from a single view without priors or special lighting conditions.

Findings

Successfully reconstructs visible and occluded geometry.

Improves generalization with low-resolution sensor data.

Operates effectively under ambient lighting conditions.

Abstract

3D reconstruction from a single-view is challenging because of the ambiguity from monocular cues and lack of information about occluded regions. Neural radiance fields (NeRF), while popular for view synthesis and 3D reconstruction, are typically reliant on multi-view images. Existing methods for single-view 3D reconstruction with NeRF rely on either data priors to hallucinate views of occluded regions, which may not be physically accurate, or shadows observed by RGB cameras, which are difficult to detect in ambient light and low albedo backgrounds. We propose using time-of-flight data captured by a single-photon avalanche diode to overcome these limitations. Our method models two-bounce optical paths with NeRF, using lidar transient data for supervision. By leveraging the advantages of both NeRF and two-bounce light measured by lidar, we demonstrate that we can reconstruct visible and occluded geometry without data priors or reliance on controlled ambient lighting or scene albedo. In addition, we demonstrate improved generalization under practical constraints on sensor spatial- and temporal-resolution. We believe our method is a promising direction as single-photon lidars become ubiquitous on consumer devices, such as phones, tablets, and headsets.