Identifying objects that are common to a set of images is an important step in unsupervised image analysis. While existing approaches to object extraction and co-segmentation commonly focus merely on joint per-class appearance models, we propose a model that incorporates additional label and boundary co-occurrence information with little additional communication across images. This model admits...

We propose the bound-optimal cutting plane method. It is a new paradigm for cutting plane generation in Mixed Integer Programming allowing for the simultaneous generation of k cuts which, when added to the current Linear Programming relaxation, yield the largest bound improvement. By Linear Programming duality arguments and standard linearization techniques we show that, for a large family of c...

We develop a method that incorporates column generation within the Lshaped method for solving two-stage stochastic linear programs. This method adds feasibility and optimality cuts generated from the second-stage subproblems based on the current columns in the master problem of the L-shaped method. It also generates columns for the master problem with respect to the first-stage constraints as w...

Graph cuts have become very popular in many areas of computer vision including segmentation, energy minimization, and 3D reconstruction. Their ability to find optimal results efficiently and the convenience of usage are some of the factors of this popularity. However, there are a few issues with graph cuts, such as inherent sequential nature of popular algorithms and the memory bloat in large s...