CRF-RNN is a formulation of a CRF as a Recurrent Neural Network. Specifically it formulates mean-field approximate inference for the Conditional Random Fields with Gaussian pairwise potentials as Recurrent Neural Networks.
Source: Conditional Random Fields as Recurrent Neural NetworksPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Semantic Segmentation | 3 | 37.50% |
Image Segmentation | 2 | 25.00% |
Brain Tumor Segmentation | 1 | 12.50% |
Tumor Segmentation | 1 | 12.50% |
Real-Time Semantic Segmentation | 1 | 12.50% |