Welcome to the Combinatorial Computer Vision group page at the Max Planck Institute for Informatics. Our goal is to develop efficient and scalable algorithms for discrete optimization problems in computer vision and beyond.
Paper accepted at ECCV 2020: Deep Graph Matching via Blackbox Differentiation of Combinatorial Solvers.
03. June 2020Paper accepted at ICML 2020: Lifted Disjoint Paths with Application in Multiple Object Tracking.
06. Januar 2020Paper accepted at AISTATS 2020: A Primal-Dual Solver for Large-Scale Tracking-by-Assignment.