Identity Inference: Generalizing Person Re-identification Scenarios Svebor Karaman and Andrew D. Bagdanov Media Integration and Communication Center University of Florence, Viale Morgagni 65, Florence, Italy svebor.karaman@unifi.it, bagdanov@dsi.unifi.it Abstract. In this article we introduce the problem of identity inference as a generalization of the re-identification problem. Identity inference is applicable in situations where a large number of unknown persons must be identified without knowing a priori that groups of test images represent the same individual. Standard single- and multi-shot person re-identification are special cases of our formulation. We present an ap- proach to solving identity inference problems using a Conditional Ran- dom Field (CRF) to model identity inference as a labeling problem in the CRF. The CRF model ensures that the final labeling gives similar la- bels to detections that are similar in feature space, and is flexible enough to incorporate constraints in the temporal and spatial domains. Exper- imental results are given on the ETHZ dataset. Our approach yields state-of-the-art performance for the multi-shot re-identification task and promising results for more general identity inference problems. 1 Introduction Person re-identification is traditionally defined as the recognition of an individual at different times, over different camera views and/or locations, and considering a large number of candidate individuals. It is a standard component of multi- camera surveillance systems as it is a way to associate multiple observations of the same individual over time. Particularly in scenarios in which the long-term behavior of persons must be characterized, accurate re-identification is essential. In realistic, wide-area surveillance scenarios such as airports, metro and train stations, re-identification systems should be capable of robustly associating a unique identity with hundreds, if not thousands, of individual observations. Re-identification performance is usually evaluated as a retrieval problem. Given a gallery consisting of a number of known individuals and images of each, for each test image or group of test images of an unknown person the goal of re-identification is to return a ranked list of individuals from the gallery. Con- figurations of the re-identification problem are generally classified according to how much group structure is available in the gallery and test image sets. In a single-shot image set there is no grouping information availabled. Though there might be multiple images of an individual, there is no knowledge of which images correspond to that person. In a multi-shot image set, on the other hand, there A. Fusiello et al. (Eds.): ECCV 2012 Ws/Demos, Part I, LNCS 7583, pp. 443–452, 2012. c  Springer-Verlag Berlin Heidelberg 2012