fork from DCurro, and modify to supprot batch size > 1
- modify netcanny.py(NOT .pyc) that support batch size >=1.
Uses PyTorch 0.4.1 and Python 3.7 (but probably works with 2.7 also).
A simple implementation of the Canny Edge Detection Algorithm (currently without hysteresis).
This project was implemented with PyTorch to take advantage of the parallelization of convolutions.
The original image:
Finding the gradient magnitude:
Early thresholding (to show that edge thingging matters):
And finally, the image after non-maximum supressions:
