cellpose-sam-service

Cellpose-SAM Detection Service

Cell segmentation service using Cellpose-SAM, running on HuggingFace ZeroGPU.

Features

• Automatic cell detection and segmentation using Cellpose-SAM
• Configurable parameters: resize, flow threshold, cell probability threshold
• Returns GeoJSON FeatureCollection with polygon geometries
• Runs on ZeroGPU (A10G) for fast inference

Interactive Use

Upload an image and adjust the detection parameters to segment cells.

API Use

from gradio_client import Client
import base64

client = Client("YOUR_USERNAME/cellpose-zero-gpu")

with open("image.png", "rb") as f:
    img_b64 = base64.b64encode(f.read()).decode()

result = client.predict(
    image_b64=img_b64,
    resize=512,
    max_iter=2000,
    flow_threshold=0.4,
    cellprob_threshold=0.0,
    api_name="/predict_api"
)