3D viewpoint-based photo search and information browsing

Kadobayashi, Rieko; Tanaka, Katsumi

doi:10.1145/1076034.1076158

Cited by 20 publications

(13 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More related to our work, Ephstein et al [10] proposed to relate images with their view frustum (viewable scene) and used a scene-centric ranking to generate a hierarchical organization of images. Several additional methods were proposed for organizing images based on camera location, direction, and additional meta-data [11][12][13][14]. Although these approaches are similar to ours in using the camera field-of-view to describe the viewable scene, their main contribution is on image browsing and grouping of similar images together.…”

Section: Related Workmentioning

confidence: 98%

Relevance ranking in georeferenced video search

Zimmermann

Kim

2010

Multimedia Systems

View full text Add to dashboard Cite

The rapid adoption and deployment of ubiquitous video cameras has led to the collection of voluminous amounts of media data. However, indexing and searching of large video databases remain a very challenging task. Recently, some recorded video data are automatically annotated with meta-data collected from various sensors such as Global Positioning System (GPS) and compass devices. In our earlier work, we proposed the notion of a viewable scene model derived from the fusion of location and direction sensor information with a video stream. Such georeferenced media streams are useful in many applications and, very importantly, they can effectively be searched via their meta-data on a large scale. Consequently, search by geo-properties complements traditional contentbased retrieval methods. The result of a georeferenced video query will in general consist of a number of video segments that satisfy the query conditions, but with more or less relevance. For example, a building of interest may appear in a video segment, but may only be visible in a corner. Therefore, an essential and integral part of a video query is the ranking of the result set according to the relevance of each clip. An effective result ranking is even more important for video than it is for text search, since the browsing of results can only be achieved by viewing each clip, which is very time consuming. In this study, we investigate and present three ranking algorithms that use spatial and temporal properties of georeferenced videos to effectively rank search results. To allow our techniques to scale to large video databases, we further introduce a histogram-based approach that allows fast online computations. An experimental evaluation demonstrates the utility of the proposed methods.

show abstract

Section: Related Workmentioning

confidence: 98%

Relevance ranking in georeferenced video search

Zimmermann

Kim

2010

Multimedia Systems

View full text Add to dashboard Cite

show abstract

“…PhotoCompas (Naaman et al 2004) clusters images based on time and location. Realityflythrough (McCurdy and Griswold 2005) uses interface ideas similar to ours for exploring video from camcorders instrumented with GPS and tilt sensors, and Kadobayashi and Tanaka (2005) present an interface for retrieving images using proximity to a virtual camera. In Photowalker (Tanaka et al 2002), a user can manually author a walkthrough of a scene by specifying transitions between pairs of images in a collection.…”

Section: Image Browsing Retrieval and Annotationmentioning

confidence: 99%

Modeling the World from Internet Photo Collections

2007

View full text Add to dashboard Cite

There are billions of photographs on the Internet, comprising the largest and most diverse photo collection ever assembled. How can computer vision researchers exploit this imagery? This paper explores this question from the standpoint of 3D scene modeling and visualization. We present structure-from-motion and image-based rendering algorithms that operate on hundreds of images downloaded as a result of keyword-based image search queries like "Notre Dame" or "Trevi Fountain." This approach, which we call Photo Tourism, has enabled reconstructions of numerous well-known world sites. This paper presents these algorithms and results as a first step towards 3D modeling of the world's well-photographed sites, cities, and landscapes from Internet imagery, and discusses key open problems and challenges for the research community.

show abstract

“…Regarding the spatial referencing of iconographic data, current research lays on manual methods based on the knowledge [Kadobayashi and Tanaka 2005] and on the registration of the position of the cameras [Waldhäusl and Ogleby 1994], on semi-automatic methods based on geometric solutions for the calibration and orientation of the cameras [Tsai 1986], or still on automatic methods based on analysis and image processing. These last methods estimate the position and orientation of the cameras, starting from the automatic extraction of vanishing points [Lee, Jung, and Nevatia 1992] or homologous points identified on the image [Snavely, Seitz, and Szeliski 2006].…”

Section: Previous Workmentioning

confidence: 99%

“…These last methods estimate the position and orientation of the cameras, starting from the automatic extraction of vanishing points [Lee, Jung, and Nevatia 1992] or homologous points identified on the image [Snavely, Seitz, and Szeliski 2006]. Some browsers have been implemented in order to manage 2D/3D datasets laying on these automatic methods: Photocloud [Brivio et al 2011] or Photo Tourism [Snavely, Seitz, and Szeliski 2007] manages photographs collections in space, while 4D Cities [Schindler, Dellaert, and Kang 2007] integrate the temporal dimension by means of a constraint satisfaction method, by ordering chronologically the spatialized photos and by automatically applying textures on 3D-models.…”

Section: Previous Workmentioning

confidence: 99%

A web platform for the consultation of spatialized and semantically enriched iconographic sources on cultural heritage buildings

Stefani¹,

Busayarat²,

Lombardo³

et al. 2013

J. Comput. Cult. Herit.

View full text Add to dashboard Cite

is an open access repository that collects the work of Arts et Métiers ParisTech researchers and makes it freely available over the web where possible. This paper describes an interactive platform for the semantic annotation of oriented iconographic sources based on an accurate 3D model structured according to spatial and temporal features. This platform, called LOCUS IMAGINIS, provides access to cultural information about monuments by collecting personal snapshots taken by visitors. In particular, the platform enables to perform two actions. Firstly, pictures are added into the database: visitors manually align images to the digital model of the monument and than, by means of automatic procedures, images are enriched with 2D semantic annotations that are projected from the 3D model. Secondly, the collection of semantically-enriched data can be queried: when visitors select 2D images of the database, these are oriented in the 3D scene and superimposed onto the 3D model. In this way, this tool provides access to textual and graphic information describing the photographed site (semantic, spatial, temporal features, etc.). Moreover, visitors can interact both with images and the 3D-model to highlight building parts by color. Accessible on-site and on-line, this tool can create awareness about conservation and is adapted to all kinds of audience: students involved in educational workshops, visitors needing to extend their visit, and documentalists interested in the consultation of the iconographic corpus.

show abstract

3D viewpoint-based photo search and information browsing

Cited by 20 publications

References 0 publications

Relevance ranking in georeferenced video search

Relevance ranking in georeferenced video search

Modeling the World from Internet Photo Collections

A web platform for the consultation of spatialized and semantically enriched iconographic sources on cultural heritage buildings

Contact Info

Product

Resources

About