Skip to content

Latest commit

 

History

History
10 lines (9 loc) · 590 Bytes

README.md

File metadata and controls

10 lines (9 loc) · 590 Bytes

Training models for 3D images

Since original dataset is not publicly available, the training pipeline based on TF Dataset API can be used and adapted to read binary data. The models operate on 3D volumetric image data in different ways, including weight-shared 2D-CNNs for sliced-input, and 3D-CNNs for a more end-to-end approach.