This document discusses techniques for interpreting point cloud and image data through automated algorithms that translate human visual interpretations. It describes popular approaches for processing LiDAR point clouds, including height-based segmentation to classify features above the ground and shape-fitting algorithms. It also discusses using spectral information through intensity values or image fusion. Finally, it examines developing "computer vision" tools that can segment data based on visual cues humans use like color, texture, morphology, context and defined shapes. The goal is to replicate human visual interpretation abilities through algorithms.