Machine learning-based edge detector.

[atonalwilson]

I have Read the following blog on edge detection using machine learning. They

used a modern machine learning-based algorithm. The algorithm is trained on images where humans annotate the most significant edges and object boundaries. Given this labeled dataset, a machine learning model is trained to predict the probability of each pixel in an image belonging to an object boundary. I would like to implement this technique using opencv. Related to this topic all issues were discussed in the machine learning online course.

Does anybody have an idea or know how this method can be implemented/developped using Opencv ?

How can we annotate the most significant edges and object boundaries for use with machine learning algorith ?