Deep Imbalanced Multi-Target Regression: 3D Point Cloud Voxel Content Estimation in Simulated Forests

2025-11-16

AI Summary

This study addresses structural information loss in forest LiDAR point cloud voxelization by proposing a deep learning framework for class-imbalanced multi-target regression, enabling high-accuracy inference of occupancy percentages—e.g., bark, leaves, soil—within fine-grained voxels from coarse (high-level) voxel inputs. Methodologically, it integrates Kernel Point Convolutions (KPConv) with a density-aware cost-sensitive learning scheme, jointly optimizing weighted mean squared error, focal regression loss, and regularization terms on DIRSIG-simulated LiDAR data. Experiments demonstrate significant mitigation of class imbalance effects; moreover, a voxel-scale accuracy trade-off is revealed: 2 m voxels yield lower overall error, whereas 0.25–0.5 m voxels exhibit elevated errors in canopy regions—highlighting the critical dependence of forest scene modeling on scale selection. To our knowledge, this work represents the first dedicated investigation of deep imbalanced multi-target regression applied to simulated forest LiDAR data.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Non-convex Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Voxelization is an effective approach to reduce the computational cost of processing Light Detection and Ranging (LiDAR) data, yet it results in a loss of fine-scale structural information. This study explores whether low-level voxel content information, specifically target occupancy percentage within a voxel, can be inferred from high-level voxelized LiDAR point cloud data collected from Digital Imaging and remote Sensing Image Generation (DIRSIG) software. In our study, the targets include bark, leaf, soil, and miscellaneous materials. We propose a multi-target regression approach in the context of imbalanced learning using Kernel Point Convolutions (KPConv). Our research leverages cost-sensitive learning to address class imbalance called density-based relevance (DBR). We employ weighted Mean Saquared Erorr (MSE), Focal Regression (FocalR), and regularization to improve the optimization of KPConv. This study performs a sensitivity analysis on the voxel size (0.25 - 2 meters) to evaluate the effect of various grid representations in capturing the nuances of the forest. This sensitivity analysis reveals that larger voxel sizes (e.g., 2 meters) result in lower errors due to reduced variability, while smaller voxel sizes (e.g., 0.25 or 0.5 meter) exhibit higher errors, particularly within the canopy, where variability is greatest. For bark and leaf targets, error values at smaller voxel size datasets (0.25 and 0.5 meter) were significantly higher than those in larger voxel size datasets (2 meters), highlighting the difficulty in accurately estimating within-canopy voxel content at fine resolutions. This suggests that the choice of voxel size is application-dependent. Our work fills the gap in deep imbalance learning models for multi-target regression and simulated datasets for 3D LiDAR point clouds of forests.

Problem

• Estimating voxel content percentages from LiDAR data in forests
• Addressing class imbalance in multi-target regression using KPConv
• Analyzing voxel size sensitivity for forest structure estimation

Innovation

• KPConv for multi-target regression on imbalanced data
• Cost-sensitive learning with density-based relevance weighting
• Sensitivity analysis of voxel size for forest structure