Tour the world: Building a web-scale landmark recognition engine

Abstract

Modeling and recognizing landmarks at world-scale is a useful yet challenging task. There exists no readily available list of worldwide landmarks. Obtaining reliable visual models for each landmark can also pose problems, and efficiency is another challenge for such a large scale system. This paper leverages the vast amount of multimedia data on the web, the availability of an Internet image search engine, and advances in object recognition and clustering techniques, to address these issues. First, a comprehensive list of landmarks is mined from two sources: (1) ∼20 million GPS-tagged photos and (2) online tour guide web pages. Candidate images for each landmark are then obtained from photo sharing websites or by querying an image search engine. Second, landmark visual models are built by pruning candidate images using efficient image matching and unsupervised clustering techniques. Finally, the landmarks and their visual models are validated by checking authorship of their member images. The resulting landmark recognition engine incorporates 5312 landmarks from 1259 cities in 144 countries. The experiments demonstrate that the engine can deliver satisfactory recognition performance with high efficiency.

Worldwide Pose Estimation Using 3D Point Clouds

“…can be answered in several ways. Some techniques approach the problem as that of classification into one of a predefined set of places (e.g., "Eiffel Tower," "Arc de Triomphe")-i.e., the "landmark recognition/classification" problem [12,13]. Other methods create a database of localized imagery and formulate the problem as one of image retrieval, after which the query image can be associated with the location of the retrieved images.…”

Location Recognition Using Prioritized Feature Matching

“…Li et al [3] derive "iconic" images derived from performing clustering on large Internet photo collections, then localize query images by retrieving similar iconic images using bag-of-words or GIST descriptors [18]. Similarly, [19] builds a landmark recognition engine by selecting iconic images using a graph-based algorithm.…”

Discovering Latent Clusters from Geotagged Beach Images

“…In particular, we use beach images in our experiments. Note that our problem is different from that of landmark recognition [23]. A landmark usually corresponds one view or one subject with a unique appearance, while a beach scene may contain a lot of clues including water, boats, people dresses, buildings and plants.…”

“…Using SVM classifiers, a novel image is geolocated by assigning it to the best cluster based on its visual content and annotations. In a recent research work [23] supported by Google, Zhen et al built a web-scale landmark recognition engine named "Tour the world" using 20 million GPS-tagged photos of landmarks together with online tour guide web pages. The experiments demonstrate that the engine can deliver satisfactory recognition performance with high efficiency.…”