Mean Field inference of CRFs based on GAT
This incremental improvement addresses inference efficiency for sequence annotation tasks like image segmentation and text annotation.
The paper tackles the problem of improving mean-field inference for fully connected pairwise CRFs by replacing linear convolution with graph attention operations, turning inference into a GAT forward process, and achieves results equivalent to a classifier with only unary potential.
In this paper we propose an improved mean-field inference algorithm for the fully connected paired CRFs model. The improved method Message Passing operation is changed from the original linear convolution to the present graph attention operation, while the process of the inference algorithm is turned into the forward process of the GAT model. Combined with the mean-field inferred label distribution, it is equivalent to the output of a classifier with only unary potential. To this end, we propose a graph attention network model with residual structure, and the model approach is applicable to all sequence annotation tasks, such as pixel-level image semantic segmentation tasks as well as text annotation tasks.