Zero-Shot Tree Detection and Segmentation from Aerial Forest Imagery

TL;DR

This paper explores the use of the Segment Anything Model 2 (SAM2) for zero-shot detection and segmentation of individual trees in aerial imagery, demonstrating its strong generalization and synergy with existing methods.

Contribution

It introduces the application of SAM2 in a zero-shot setting for tree detection and segmentation in remote sensing imagery, highlighting its generalization and potential for combining with specialized models.

Findings

SAM2 exhibits impressive zero-shot generalization to tree segmentation.

Combining SAM2 with existing detection models enhances segmentation accuracy.

Large pretrained models are promising for remote sensing tasks.

Abstract

Large-scale delineation of individual trees from remote sensing imagery is crucial to the advancement of ecological research, particularly as climate change and other environmental factors rapidly transform forest landscapes across the world. Current RGB tree segmentation methods rely on training specialized machine learning models with labeled tree datasets. While these learning-based approaches can outperform manual data collection when accurate, the existing models still depend on training data that's hard to scale. In this paper, we investigate the efficacy of using a state-of-the-art image segmentation model, Segment Anything Model 2 (SAM2), in a zero-shot manner for individual tree detection and segmentation. We evaluate a pretrained SAM2 model on two tasks in this domain: (1) zero-shot segmentation and (2) zero-shot transfer by using predictions from an existing tree detection model as prompts. Our results suggest that SAM2 not only has impressive generalization capabilities, but also can form a natural synergy with specialized methods trained on in-domain labeled data. We find that applying large pretrained models to problems in remote sensing is a promising avenue for future progress. We make our code available at: https://github.com/open-forest-observatory/tree-detection-framework.