---
license: apache-2.0
language:
- en
tags:
- cell-segmentation
- biology
- microscopy
- sam2
- biomedical-imaging
- instance-segmentation
base_model: facebook/sam2.1-hiera-tiny
datasets:
- DnaRnaProteins/cell_seg_labeled
pipeline_tag: image-segmentation
---

# SAM2-tiny — Cell Segmentation

Fine-tuned [SAM2-tiny](https://huggingface.co/facebook/sam2.1-hiera-tiny) for
instance segmentation of cells in fluorescence microscopy images.
Part of the [biomech-inference-serving](https://github.com/czi-ai/biomech-inference-serving-feedback)
pipeline (internal research project).

## Training

| | |
|---|---|
| Base model | `facebook/sam2.1-hiera-tiny` |
| Training data | [`DnaRnaProteins/cell_seg_labeled`](https://huggingface.co/datasets/DnaRnaProteins/cell_seg_labeled) |
| Fine-tuning | Full decoder fine-tune |
| Framework | [sam2](https://github.com/facebookresearch/sam2) |

## Usage

```python
import numpy as np, torch
from PIL import Image
from sam2.sam2_image_predictor import SAM2ImagePredictor

predictor = SAM2ImagePredictor.from_pretrained("DnaRnaProteins/sam2-cells-seg")

image = np.array(Image.open("cell_image.png").convert("RGB"))
predictor.set_image(image)

with torch.inference_mode():
    masks, scores, _ = predictor.predict(
        point_coords=np.array([[128, 256]]),  # [x, y] prompt point
        point_labels=np.array([1]),
        multimask_output=True,
    )
# masks: (N, H, W) bool array
# scores: (N,) float confidence per mask
```

## Via Modal endpoint

```python
import base64, modal

segment = modal.Function.from_name("biomech-inference-serving", "segment")
with open("cell_image.png", "rb") as f:
    b64 = base64.b64encode(f.read()).decode()
result = segment.remote(b64)
# {"masks": [[...]], "scores": [0.94, ...]}
```

## Limitations

- Optimised for fluorescence cell images; performance on brightfield or H&E may vary.
- Point prompts improve precision — promptless predictions use a default center point.