Probabilistic Correlation Clustering and Image Partitioning Using Perturbed Multicuts
- Cite this paper as:
- Kappes J.H., Swoboda P., Savchynskyy B., Hazan T., Schnörr C. (2015) Probabilistic Correlation Clustering and Image Partitioning Using Perturbed Multicuts. In: Aujol JF., Nikolova M., Papadakis N. (eds) Scale Space and Variational Methods in Computer Vision. SSVM 2015. Lecture Notes in Computer Science, vol 9087. Springer, Cham
We exploit recent progress on globally optimal MAP inference by integer programming and perturbation-based approximations of the log-partition function. This enables to locally represent uncertainty of image partitions by approximate marginal distributions in a mathematically substantiated way, and to rectify local data term cues so as to close contours and to obtain valid partitions. Our approach works for any graphically represented problem instance of correlation clustering, which is demonstrated by an additional social network example.
KeywordsCorrelation clustering Multicut Perturb and MAP
Unable to display preview. Download preview PDF.