ForestQuery: Boundary-Aware and Spatially Anchored Query Learning for Unified Forest Point Cloud Segmentation

2026-10-02

AI Summary

This study addresses the challenge of joint semantic and instance segmentation in forest point clouds, which is hindered by irregular structures, occlusion, and ambiguous boundaries. To this end, we propose ForestQuery, a framework grounded in a query learning paradigm. It optimizes query construction by explicitly modeling boundary uncertainty and introduces a spatially anchored semantic enhancement mechanism that integrates learnable 3D anchors with vertical stratification priors into feature encoding. Furthermore, an adaptive loss reweighting strategy is employed to improve training stability. Extensive experiments demonstrate that the proposed framework significantly enhances both individual tree instance and semantic segmentation performance across multiple public benchmarks and real-world datasets. The source code and associated data have been made publicly available.

Abstract

Forest point cloud segmentation is fundamental for fine-grained 3D forest scene understanding, yet remains challenging due to irregular tree structures, severe occlusions, density variations, and ambiguous instance boundaries. Recent query-based forest segmentation methods have shown promise for unified semantic and instance prediction, but they still insufficiently exploit forest-specific spatial structure and account for boundary uncertainty. In this paper, we propose ForestQuery, a boundary-aware and spatially anchored query learning framework for unified forest point cloud segmentation. ForestQuery enhances instance and semantic query learning through two complementary designs. Specifically, boundary uncertainty is explicitly modeled to guide reliable instance query construction and modulate query optimization through adaptive loss reweighting. Meanwhile, spatially anchored semantic query enhancement (SA-SQE) introduces learnable 3D anchors encoding forest vertical stratification priors to enrich semantic queries with explicit spatial references. We evaluate ForestQuery on multiple public forest point cloud benchmarks and a self-collected annotated real-world dataset. Extensive experiments demonstrate consistent improvements in both individual-tree segmentation and semantic segmentation across diverse forest scenes. Code and data are publicly available at https://zhan994.github.io/ForestQuery

Problem

• forest point cloud segmentation
• instance boundary ambiguity
• boundary uncertainty
• spatial structure
• unified segmentation

Innovation

• Forest Point Cloud Segmentation
• Boundary-Aware Query Learning
• Spatially Anchored Semantic Query Enhancement
• Unified Segmentation
• 3D Anchors