Assessing SAM for Tree Crown Instance Segmentation from Drone Imagery
This work addresses the problem of inefficient tree monitoring for climate solutions by testing advanced vision models, but it is incremental as it shows limited immediate gains over existing methods.
The study evaluated the Segment Anything Model (SAM) for automatic tree crown instance segmentation in drone imagery of young plantations, finding that out-of-the-box SAM methods did not outperform a custom Mask R-CNN, but improvements were possible with further tuning and adding Digital Surface Model (DSM) information.
The potential of tree planting as a natural climate solution is often undermined by inadequate monitoring of tree planting projects. Current monitoring methods involve measuring trees by hand for each species, requiring extensive cost, time, and labour. Advances in drone remote sensing and computer vision offer great potential for mapping and characterizing trees from aerial imagery, and large pre-trained vision models, such as the Segment Anything Model (SAM), may be a particularly compelling choice given limited labeled data. In this work, we compare SAM methods for the task of automatic tree crown instance segmentation in high resolution drone imagery of young tree plantations. We explore the potential of SAM for this task, and find that methods using SAM out-of-the-box do not outperform a custom Mask R-CNN, even with well-designed prompts, but that there is potential for methods which tune SAM further. We also show that predictions can be improved by adding Digital Surface Model (DSM) information as an input.