The SingleTree project, in collaboration with the Swedish University of Agricultural Sciences (SLU), is thrilled to announce the launch of a new Benchmark Challenge in Forest Sensing and Algorithms. Deadline for submissions 31 st December 2026.
This competition invites researchers, data scientists, and developers from around the world to test and evaluate their models against real-world forest data. The primary goal? To gather and evaluate sensor- and platform-agnostic algorithms for whorl detection and stem profile estimation.
Driving the Future of Precision Forestry
This benchmark is not just a technical competition; it is a vital step for the technological development of the SingleTree project. By evaluating the performance of these algorithms, we aim to:
- Understand the potential and limitations of current state-of-the-art algorithms in forestry tech.
- Demonstrate practical applications: Show how detailed information can be implemented to transform the value chain and manage forests at the single-tree level, both today and in the future.
- Generate actionable insights: Automatically derive new tree quality variables from remote sensing data. This information is crucial for better categorizing timber assortments, which directly enhances forest operations such as bucking optimization and timber flow.
The Challenge: Dataset and Tasks
Participants will use the FOR-wood dataset, which is openly available for download from Zenodo. Competitors are required to use their models on the provided test set to predict either:
- Whorl-positions relative to the ground.
- The stem profile diameter.
- Or both.
Predictions must be submitted in specific .csv formats, evaluated against metrics like F1 (for whorl detection) and RMSE (for stem profile estimation) on our live leaderboard.
The Prize: Co-author a Scientific Publication
We believe in collaborative science. The top five participating teams will earn the opportunity to co-author an official scientific article about this benchmark. Each winning team can nominate up to two people to be registered as co-authors.
This paper will be published officially within the framework of the SingleTree project, featuring the corresponding European Union funding acknowledgments, offering participants significant visibility in the scientific and forestry communities.
How to Participate
Are you ready to shape the future of forest technology?
- Read the Terms: Go to the terms tab under “Get Started” on the submission platform to ensure compliance.
- Download the Dataset: The FOR-wood dataset is available on Zenodo: https://zenodo.org/records/23012506https://lnkd.in/dBWdcEY8
- Submit Your Predictions: You can submit up to 2 times per day to refine your models. Access the submission portal here: https://www.codabench.org/competitions/18295/
