Skip to content

Latest commit

 

History

History
40 lines (32 loc) · 1.08 KB

README.md

File metadata and controls

40 lines (32 loc) · 1.08 KB

tf-SegCaps

TensorFlow implementation of SegCaps [1]
I'm now testing this implementation with ISBI2012. (in progress)

model_figure

Requirements

  • python 2.7
  • Tensorflow >= 1.4
  • numpy
  • scipy
  • skimage
  • cv2

Usage

Before use, first check options in config.py

Options
  batch_size : The number examples in a batch
  max_iter   : The number of iteration of training
  mask       : Whether to use mask in reconsturction loss.
  mask_inv   : Whether to apply color inversion to the image intensity.
  • trainining: python train.py --device 0 --mask True --mask_inv True
  • deployment: python deploy.py --device 0 --mask True --mask_inv True

Result

  • ISBI 2012 dataset

training_curve
example_test example_result
(The result seems too much overfitted, I'm now testing the several data augmentations.)

Reference

[1] Capsules for Object Segmentation (https://arxiv.org/abs/1804.04241)

Authors

Inwan Yoo / [email protected]