FastGeodis icon indicating copy to clipboard operation
FastGeodis copied to clipboard

Fast Implementation of Generalised Geodesic Distance Transform for CPU (OpenMP) and GPU (CUDA)

Results 5 FastGeodis issues
Sort by recently updated
recently updated
newest added

`ValueError: FastGeodis currently only supports single batch input.` when using `FastGeodis.generalised_geodesic2d, v=1.0e10, lamb=0.0, iter=1` - Is there any timeline on implementing multiple batch input - Currently I run each batch...

**Is your feature request related to a problem? Please describe.** The package currently provide GPU-based approximations of Geodesic and Euclidean distance transforms. Following #42 fast marching now allows for more...

**Is your feature request related to a problem? Please describe.** The cuda-version generalised_geodesic3d has shown non-deterministic behaviour. Given the same inputs, it returns different output (geodesic map) at different runs....

**Describe the bug** Clean up and improve code for pixel queue method as described in the comment [here](https://github.com/masadcv/FastGeodis/pull/45#issuecomment-1494779849): > For clarity, it may be good to clean up a bit...

If a system has pytorch pre-installed using conda - then poetry (pyproject.toml) does not detect it and installs pytorch using pip This causes issues especially if there is a mismatch...