This chapter presents distributed algorithms for solving two classical computer vision problems: object localization, i.e., estimation of the 3-D pose of an object from multiple 2-D views; and camera sensor network localization, i.e., estimation of the 3-D camera poses from multiple 2-D views. These problems are posed as generalized consensus problems in the space of rigid-body motions and solved using extensions of classical consensus algorithms to this space. The convergence properties of the proposed algorithms are studied and compared to naïve generalizations of Euclidean consensus, which fail to converge to the correct solution. Experiments on synthetic data evaluate the proposed methods.