NeRF-Accelerated Ecological Monitoring in Mixed-Evergreen Redwood Forest

TL;DR

This paper compares mobile laser scanning and neural radiance fields for forest mapping, introducing a convex-hull method that improves tree diameter estimation accuracy in Redwood forests.

Contribution

It presents a novel comparison between MLS and NeRF reconstructions for forest mapping and introduces an improved DBH estimation method using convex-hull modeling.

Findings

NeRF-based reconstructions outperform traditional methods in accuracy.

Convex-hull modeling reduces DBH estimation RMSE to 1.68 cm.

Code and datasets are publicly available for further research.

Abstract

Forest mapping provides critical observational data needed to understand the dynamics of forest environments. Notably, tree diameter at breast height (DBH) is a metric used to estimate forest biomass and carbon dioxide sequestration. Manual methods of forest mapping are labor intensive and time consuming, a bottleneck for large-scale mapping efforts. Automated mapping relies on acquiring dense forest reconstructions, typically in the form of point clouds. Terrestrial laser scanning (TLS) and mobile laser scanning (MLS) generate point clouds using expensive LiDAR sensing, and have been used successfully to estimate tree diameter. Neural radiance fields (NeRFs) are an emergent technology enabling photorealistic, vision-based reconstruction by training a neural network on a sparse set of input views. In this paper, we present a comparison of MLS and NeRF forest reconstructions for the purpose of trunk diameter estimation in a mixed-evergreen Redwood forest. In addition, we propose an improved DBH-estimation method using convex-hull modeling. Using this approach, we achieved 1.68 cm RMSE, which consistently outperformed standard cylinder modeling approaches. Our code contributions and forest datasets are freely available at https://github.com/harelab-ucsc/RedwoodNeRF.