Unsupervised learning by probabilistic latent semantic analysis
Machine Learning
The Journal of Machine Learning Research
Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
A Bayesian Hierarchical Model for Learning Natural Scene Categories
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
A Performance Evaluation of Local Descriptors
IEEE Transactions on Pattern Analysis and Machine Intelligence
Discovering Objects and their Localization in Images
ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1 - Volume 01
Photo tourism: exploring photo collections in 3D
ACM SIGGRAPH 2006 Papers
Remote Sensing and Image Interpretation
Remote Sensing and Image Interpretation
Modeling Semantic Aspects for Cross-Media Image Indexing
IEEE Transactions on Pattern Analysis and Machine Intelligence
Generating diverse and representative image search results for landmarks
Proceedings of the 17th international conference on World Wide Web
Geotagged iage rcognition by cmbining tree dfferent knds of golocation fatures
ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part II
Identification of scene locations from geotagged images
ACM Transactions on Multimedia Computing, Communications, and Applications (TOMCCAP)
| Hi-index | 0.00 |
In this paper, we propose to exploit geotags as additional information for visual recognition of consumer photos to improve its performance. Geotags, which represent places where the photos were taken, for photos can be obtained automatically by carrying a portable small GPS device with digital cameras. Geotags have potential to improve performance of visual image recognition, since recognition targets are unevenly distributed. For example, "beach" photos can be taken near the sea and "lion" photos can be taken only in a zoo except Africa. To integrate geotag information into visual image recognition, we adopt two types of geographical information, raw values of latitude and longitude, and visual feature of aerial photos around the location the geotag represents. As classifiers, we use both a discriminative method and a generative method in the experiments. The objective of this paper is to examine if geotags can help category-level image recognition. Note that we define an image recognition problem as deciding if an image is associated with a certain given concept such as "mountain" and "beach" in this paper. We propose a novel method to carry out geotagged image recognition in this paper. The experimental results demonstrate effectiveness of usage of geographical information for recognition of consumer photos.