抄録
In this paper, we address the challenging problem of single-view cross-season place recognition. A new approach is proposed for compact discriminative scene descriptor that facilitates coping with changes in appearance in the environment. We focus on a simple effective strategy that uses those landmarks whose appearance have not actually changed across seasons as valid landmarks. Unlike popular bag-of-words scene descriptors which rely on a library of vector quantized visual features, our descriptor is based on a library of raw image data, such as publicly available photo collections from Google StreetView, and directly mines it to discover landmarks (i.e., image patches) that effectively explain an input query/database image. Discovered landmarks are then compactly described by their pose and shape (i.e., library image ID, bounding boxes) and used as a compact discriminative scene descriptor for the input image. We collected a challenging dataset of single-view images across seasons with annotated ground truth, and evaluated the effectiveness of our scene description framework by comparing its performance to that of previous approaches.
