Learning and calibrating per-location classifiers for visual place recognition

International audienceThe aim of this work is to localize a query photograph by finding other images depicting the same place in a large geotagged image database. This is a challenging task due to changes in viewpoint, imaging conditions and the large size of the image database. The contribution of this work is twofold. First, we cast the place recognition problem as a classification task and use the available geo-tags to train a classifier for each location in the database in a similar manner to per-exemplar SVMs in object recognition. Second, as only one or a few positive training examples are available for each location, we propose two methods to calibrate all the per-location SVM classifiers without the need for additional positive training data. The first method relies on p-values from statistical hypothesis testing and uses only the available negative training data. The second method performs an affine calibration by appropriately normalizing the learnt classifier hyperplane and does not need any additional labelled training data. We test the proposed place recognition method with the bag-of-visual-words and Fisher vector image representations suitable for large scale indexing. Experiments are performed on three datasets: 25,000 and 55,000 geotagged street view images of Pittsburgh, and the 24/7 Tokyo benchmark containing 76, 000 images with varying illumination conditions. The results show improved Query image Retrieved locations Calibrated e-SVMs classifiers Fig. 1: The goal of this work is to localize a query photograph (left top) by finding other images of the same place in a large geotagged image database (right column). We cast the problem as a classification task and learn a classifier for each location in the database. We develop two procedures to calibrate the outputs of the large number of per-location classifiers without the need for additional labeled training data. place recognition accuracy of the learnt image representation over direct matching of raw image descriptors