Abstract
The growing explosion in the use of surveillance cameras in public security highlights the importance of vehicle search from a large-scale image or video database. However, compared with person re-identification or face recognition, vehicle search problem has long been neglected by researchers in vision community. This paper focuses on an interesting but challenging problem, vehicle re-identification (a.k.a precise vehicle search). We propose a Deep Relative Distance Learning (DRDL) method whichexploits a two-branch deep convolutional network to projectraw vehicle images into an Euclidean space where distancecan be directly used to measure the similarity of arbitrarytwo vehicles. To further facilitate the future research onthis problem, we also present a carefully-organized large-scale image database “VehicleID”, which includes multi-ple images of the same vehicle captured by different real-world cameras in a city. We evaluate our DRDL method onour VehicleID dataset and another recently-released vehi-cle model classification dataset “CompCars” in three setsof experiments: vehicle re-identification, vehicle model ver-ification and vehicle retrieval. Experimental results showthat our method can achieve promising results and outper-forms several state-of-the-art approaches.