DINOv3
Paper • 2508.10104 • Published • 319
You need to agree to share your Hugging Face information to access this model
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
Access is approved automatically. By requesting access, you share your Hugging Face username and email with HistoForge. We may get in touch to ask for your feedback.
Log in or Sign Up to review the conditions and access this model content.
DINOv3 Cellpose Starter
A starting model for training HistoForge to separate nuclei, cells and other individual objects, including irregular or elongated shapes. It uses Cellpose's flow-based method.
It starts from the official DINOv3 ViT-L/16 image encoder, with newly initialized segmentation layers that need training.
Compare starter models
H&E-12 starters include histology pretraining. DINOv3 starters use a pretrained image encoder with newly initialized segmentation layers.
| Method | Start here for | H&E-12 weights | DINOv3 weights |
|---|---|---|---|
| Cellpose | Nuclei, cells and irregular or elongated objects | H&E-12 Cellpose | DINOv3 Cellpose |
| HoVer | Large individual tissue structures | H&E-12 HoVer | DINOv3 HoVer |
| Dense semantic | Tissue compartments, rather than separate cells | H&E-12 Semantic | DINOv3 Semantic |
Cellpose and HoVer can separate touching nuclei. These are starting recommendations, not an accuracy or speed ranking; compare trained results on held-out images.
Using the module
Install histoforge/starter-dinov3-cellpose and select it as the starting model when training from your QuPath annotations. Define your target classes and image resolution, train, then validate the resulting model on held-out images before using its predictions.
| Output class | What it marks |
|---|---|
| Object | Individual segmented objects |
This starter has no instance-classification head. Classification can be added when defining your training task.
Measurements
No task-specific measurements are included.
Limitations
- Objects spanning many pixels may need more Cellpose flow iterations to form complete objects. HistoForge fixes this inference setting internally; increasing training epochs does not change it.
- The adapter and flow head are untrained; predictions are not meaningful before fine-tuning.
- This starter has no task-specific accuracy benchmark or prediction gallery.
- Input images must be RGB with valid pixel-size calibration. The initial recipe uses 0.25 microns per pixel; choose the resolution appropriate for your training data.
Models and training
| Component | Initialization |
|---|---|
| DINOv3 ViT-L/16 backbone | Official LVD-1689M weights |
| Spatial adapter and feature readout | Newly initialized |
| Task head | Newly initialized flow and foreground head |
The model uses the shared 256-channel spatial adapter and four CNN-transformer interaction blocks. Newly initialized components use seed 42. Neural tiles are 256 pixels with 10% overlap; the complete settings are in recipe.json.
The backbone is the official DINOv3 ViT-L/16 LVD-1689M model. No H&E-12 weights are used.
Licences and citations
- Weights: DINOv3 licence.
- Foundation model: Simeoni et al., DINOv3 (2025).
Model tree for histoforge/starter-dinov3-cellpose
Base model
timm/vit_large_patch16_dinov3.lvd1689m