This talk will focus on the problem of learning similarity-preserving binary codes for indexing and search of large-scale image collections. The idea is to map high-dimensional feature vectors to compact bit strings such that vectors representing semantically or perceptually similar images in the original feature space map to strings that have a low Hamming distance. I will describe two methods for learning binary codes that I have developed together with my students and collaborators, and show an application of these codes to clustering and reconstruction of landmark and city photo collections.