You can deploy this model and start earning money today!

jannikend/
dinovtree

tree-height-estimation

species-identification

fine-grained-classification

biodiversity

2603.23669

other

Views

No views yet

[ECCV 2026] DINOvTree 🦖🌳

ArXiv

Benchmark

Code

This repository contains the model weights for the DINOvTree model introduced in our paper accepted at ECCV 2026:

Estimating Individual Tree Height and Species from UAV Imagery

Authors: Jannik Endres, Etienne Laliberté, David Rolnick, Arthur Ouaknine

Our model, DINOvTree, leverages a Vision Foundation Model (VFM) to extract features and predicts the height and species of the center tree in the input image with two separate heads.

Note: For full installation, training, and evaluation instructions, please refer to our GitHub repository.

📦 Model Weights & Datasets

We publish the model weights of DINOvTree in Base size for three distinct datasets.

Dataset	Weights File	Description
Quebec Trees	`dinovtreeb_quebectrees.pth`	Temperate Forest
BCI	`dinovtreeb_bci.pth`	Tropical Forest
Quebec Plantations	`dinovtreeb_quebecplantations.pth`	Boreal Plantation

⚖️ License

The VFM of our checkpoint was initialized with DINOv3 weights, which are licensed under the DINOv3 License. The heads and our changes to the VFM weights are licensed under the Apache License 2.0.

📚 Citation

If you find our work useful, please consider citing our paper:

bibtex

@inproceedings{endres2026treeheightspecies,
  title     = {Estimating Individual Tree Height and Species from UAV Imagery},
  author    = {Endres, Jannik and Lalibert{\'e}, Etienne and Rolnick, David and Ouaknine, Arthur},
  booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
  year      = {2026}
}

jannikend/dinovtree

[ECCV 2026] DINOvTree 🦖🌳

📦 Model Weights & Datasets

⚖️ License

📚 Citation

jannikend/
dinovtree