https://www.kaggle.com/c/sartorius-cell-instance-segmentation
Deep Watershed Transform Network:
python seggit/data/scripts/make_semseg_target.py
python seggit/training/run_segmentation.py
from seggit.cell_semantic_segmentation import SemanticSegmenter
segmenter = SemanticSegmenter(checkpoint_path='best.pth')
img, semseg = segmenter.predict('sample.png')
python seggit/data/scripts/make_uvec.py
python training/run_direction.py
python seggit/data/scripts/make_wngy.py
python training/run_energy.py
python training/run_watershed.py
from seggit.deep_watershed_transform import DeepWatershedTransform
dwt = DeepWatershedTransform(checkpoint_path='best.pth') wngy = dwt.predict(img, semg)
## Cell instance segmentation (Unet + WN)
To make an inference :
from seggit.cell_instance_segmentation import CellSegmenter
parser = argparse.ArgumentParser() CellSegmenter.add_argparse_args(parser) args = parser.parse_args() args.pth_unet = 'best_unet.pth' args.pth_wn = 'best_wn.pth'
segmenter = CellSegmenter(args)
img, instg = segmenter.predict('sample.png')
# References
- [[ods.ai] topcoders, 1st place solution](https://www.kaggle.com/c/data-science-bowl-2018/discussion/54741)
- [@hengck23 [placeholder] my approach and results](https://www.kaggle.com/c/sartorius-cell-instance-segmentation/discussion/285516)
- [Deep Watershed Transform for Instance Segmentation](https://arxiv.org/pdf/1611.08303.pdf)
- https://github.com/min2209/dwt
- Segmentation Models Pytorch https://github.com/qubvel/segmentation_models.pytorch
- https://en.wikipedia.org/wiki/Distance_transform
- https://stackoverflow.com/questions/61716670/distance-transform-the-function-does-not-work-properly
- https://stackoverflow.com/questions/61204462/error-in-function-distancetransform-python-using-opencv-3-4-9
- https://github.com/MouseLand/cellpose
- https://github.com/YukangWang/TextField