Remote sensing technologies using Uncrewed Aerial Vehicles (UAVs) offers a promising approach to estimate parameters for forest inventory. Studies have used UAV laser scanning data to indirectly estimate Above Ground Biomass (AGB) or related forest features at local scales, often relying on features such as tree height, crown diameter, or intensity. However, common approaches underutilize the full potential of 3D point cloud data, which can enable direct estimation of Diameter at Breast Height (DBH), and therefore AGB. This chapter summarizes state-of-the-art methods for estimating DBH and showcases an implementation of the main ideas: provided that sufficient trunk information is captured, aerial data can be used to accurately determine DBH. The presented methodology uses two public datasets: FOR-Instance and Weiser. The former is used to develop a coarse-to-fine trunk segmentation approach, while the latter is used to assess three machine learning regressors for DBH estimation based on trunk features. Results demonstrate that UAV-based tree point clouds with a trunk point density higher than 100 pts m \(^{-2}\) are suitable for segmenting the trunk. Direct DBH computation yields a RMSE ranging from 11.9 to 17.9 cm, which can be improved by neural network regressors, reducing the Root Mean Squared Error, RMSE, to 3.8 cm, on average.