StarDist

class byotrack.implementation.detector.stardist.StarDistDetector(model: StarDist2D, **kwargs)

Bases: BatchDetector

Runs stardist as a detector.

Wraps the official implementation at https://github.com/stardist/stardist, following the paper: Uwe Schmidt, Martin Weigert, Coleman Broaddus, and Gene Myers. Cell Detection with Star-convex Polygons. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Granada, Spain, September 2018.

We do not provide any code to train the stardist model. You can only use a trained or pretrained model.

Note

This module requires stardist lib to be installed (with tensorflow). Please follow the instruction of the official implementation to install it.

model

Underlying StarDist model

Type:: StarDist2D

prob_threshold

Threshold on probability

Type:: float

nms_threshold

Threshold for Non Maximum Suppression

Type:: float

detect(batch: ndarray) → List[Detections]

Apply the detection on a batch of frames

By default, the frame ids are set from 0 to n-1 with n the size of the batch. The aggregattion of batches and frame ids correction is automatically handled when called the run method.

Parameters:: batch (np.ndarray) – Batch of video frames Shape: (B, H, W, C)
Returns:: Detections for each given frame
Return type:: Sequence[byotrack.Detections]

static from_pretrained(name: str, **kwargs) → StarDistDetector

Load a pretrained StarDist from the paper

Parameters:

name (str) – A valid identifier (From the official github)
**kwargs – Additional detector arguments. (See byotrack.BatchDetector)

Returns:

StarDistDetector

static from_trained(train_dir: str | PathLike, **kwargs) → StarDistDetector

Load a trained StarDist from a local folder

Parameters:

train_dir (str | os.PathLike) – The training folder of the model
**kwargs – Additional detector arguments. (See byotrack.BatchDetector)

Returns:

StarDistDetector