I am currently trying to add a CRF to the output of the mask graph to improve segmentation task after I saw this implementation of the referring paper: [https://github.com/Gurupradeep/FCN-for-Semantic-Segmentation]() However, I am not sure about how to get the ouput masks, and if I can load trained weights once I add this layer.
Does anyone have any idea about it?
Many thanks for this great project !
Edit : I tried non trained dense CRF but gets poor results, I am looking for the trainable version of it
Hello everyone,
I am currently trying to add a CRF to the output of the mask graph to improve segmentation task after I saw this implementation of the referring paper: [https://github.com/Gurupradeep/FCN-for-Semantic-Segmentation]() However, I am not sure about how to get the ouput masks, and if I can load trained weights once I add this layer. Does anyone have any idea about it?
Many thanks for this great project !
Edit : I tried non trained dense CRF but gets poor results, I am looking for the trainable version of it