Abstract
Metric learning methods, for person re-identi?cation, es timate a scaling for distances in a vector space that is optimized for picking out observations of the same individual. This paper presents a novel approach to the pedestrian re-identi?cation problem that uses metric learning to improve the state-of-the-art performance on standard public datasets. Very high dimensional features are extracted from the source color image. A ?rst processing stage performs unsupervised PCA dimensionality reduction, constrained to maintain the redundancy in color-space representation. A second stage further reduces the dimensionality, using a Local Fisher Discriminant Analysis de?ned by a training set. A regularization step is introduced to avoid singular matrices during this stage. The experiments conducted on three publicly available datasets con?rm that the proposed method outperforms the state-of-the-art performance, including all other known metric learning methods. Furthermore, the method is an effective way to process observations comprising multiple shots, and is non-iterative: the computation times are relatively modest. Finally, a novel statistic is derived to characterize the Match Characteristic: the normalized entropy reduction can be used to de?ne the ’Proportion of Uncertainty Removed’ (P UR). This measure is invariant to test set size and provides an intuitive indication of performance.