M. Pawan Kumar 
HOME

EFFICIENT FRANKWOLFE FOR DENSE CRFs AND PIECEWISE LINEAR CNNs M. Pawan Kumar PreDoc Summer School, ETH Zurich, 2017
The course consists of two parts. In the first part, we will consider the challenging problem of energy minimization in dense CRFs, which is typically addressed using the meanfield algorithm. The popularity of meanfield is largely due to the fact that each iteration of the algorithm can be performed efficiently using a filtering method based on the permutohedral lattice. We will consider several continuous relaxations for the problem, which provide strong theoretical guarantees on the quality of the solution, and show that they can also be solved efficiently by leveraging the power of the FrankWolfe algorithm. Specifically, each iteration of FrankWolfe can also be performed efficiently using the same filtering method.
Topic 1: Dense CRF [PPT] [PDF] 