SelvaMask: Segmenting Trees in Tropical Forests and Beyond

TL;DR

SelvaMask introduces a comprehensive tropical forest dataset and a novel detection-segmentation pipeline that significantly improves tree crown segmentation accuracy, advancing forest monitoring and ecological research.

Contribution

The paper presents a new tropical forest dataset with detailed annotations and a modular vision foundation model-based pipeline that achieves state-of-the-art segmentation performance.

Findings

Outperforms existing models in tropical forest tree crown segmentation

Demonstrates generalization to tropical and temperate datasets

Provides a publicly available dataset and code for future research

Abstract

Tropical forests harbor most of the planet's tree biodiversity and are critical to global ecological balance. Canopy trees in particular play a disproportionate role in carbon storage and functioning of these ecosystems. Studying canopy trees at scale requires accurate delineation of individual tree crowns, typically performed using high-resolution aerial imagery. Despite advances in transformer-based models for individual tree crown segmentation, performance remains low in most forests, especially tropical ones. To this end, we introduce SelvaMask, a new tropical dataset containing over 8,800 manually delineated tree crowns across three Neotropical forest sites in Panama, Brazil, and Ecuador. SelvaMask features comprehensive annotations, including an inter-annotator agreement evaluation, capturing the dense structure of tropical forests and highlighting the difficulty of the task. Leveraging this benchmark, we propose a modular detection-segmentation pipeline that adapts vision foundation models (VFMs), using domain-specific detection-prompter. Our approach reaches state-of-the-art performance, outperforming both zero-shot generalist models and fully supervised end-to-end methods in dense tropical forests. We validate these gains on external tropical and temperate datasets, demonstrating that SelvaMask serves as both a challenging benchmark and a key enabler for generalized forest monitoring. Our code and dataset will be released publicly.