Graph Cuts for Supervised Binary Coding

Tiezheng Ge¹, Kaiming He², and Jian Sun²

¹University of Science and Technology of China, China

²Microsoft Research, USA

Abstract. Learning short binary codes is challenged by the inherent discrete nature of the problem. The graph cuts algorithm is a well-studied discrete label assignment solution in computer vision, but has not yet been applied to solve the binary coding problems. This is partially because it was unclear how to use it to learn the encoding (hashing) functions for out-of-sample generalization. In this paper, we formulate supervised binary coding as a single optimization problem that involves both the encoding functions and the binary label assignment. Then we apply the graph cuts algorithm to address the discrete optimization problem involved, with no continuous relaxation. This method, named as Graph Cuts Coding (GCC), shows competitive results in various datasets.

LNCS 8695, p. 250 ff.

Full article in PDF | BibTeX

lncs@springer.com
© Springer International Publishing Switzerland 2014