Discovering objects and their location in images Export

@inproceedings{citeulike:1321552, abstract = {We seek to discover the object categories depicted in a set of unlabelled images. We achieve this using a model developed in the statistical text literature: probabilistic latent semantic analysis (pLSA). In text analysis, this is used to discover topics in a corpus using the bag-of-words document representation. Here we treat object categories as topics, so that an image containing instances of several categories is modeled as a mixture of topics. The model is applied to images by using a visual analogue of a word, formed by vector quantizing SIFT-like region descriptors. The topic discovery approach successfully translates to the visual domain: for a small set of objects, we show that both the object categories and their approximate spatial layout are found without supervision. Performance of this unsupervised method is compared to the supervised approach of Fergus et al. (2003) on a set of unseen images containing only one object per image. We also extend the bag-of-words vocabulary to include 'doublets' which encode spatially local co-occurring regions. It is demonstrated that this extended vocabulary gives a cleaner image segmentation. Finally, the classification and segmentation methods are applied to a set of images containing multiple objects per image. These results demonstrate that we can successfully build object class models from an unsupervised analysis of images.}, author = {Sivic, J. and Russell, B. C. and Efros, A. A. and Zisserman, A. and Freeman, W. T.}, booktitle = {2005 IEEE International Conference on Computer Vision (ICCV)}, citeulike-article-id = {1321552}, citeulike-linkout-0 = {http://doi.ieeecomputersociety.org/10.1109/ICCV.2005.77}, citeulike-linkout-1 = {http://dx.doi.org/10.1109/ICCV.2005.77}, citeulike-linkout-2 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1541280}, citeulike-linkout-3 = {http://www.robots.ox.ac.uk/~vgg/publications/papers/sivic05b.pdf}, doi = {10.1109/ICCV.2005.77}, journal = {Computer Vision, 2005. ICCV 2005. Tenth IEEE International Conference on}, keywords = {education}, pages = {370--377}, posted-at = {2008-11-13 06:40:48}, priority = {2}, title = {Discovering objects and their location in images}, url = {http://dx.doi.org/10.1109/ICCV.2005.77}, volume = {1}, year = {2005} }

TY - CONF ID - citeulike:1321552 L3 - citeulike-article-id:1321552 N2 - We seek to discover the object categories depicted in a set of unlabelled images. We achieve this using a model developed in the statistical text literature: probabilistic latent semantic analysis (pLSA). In text analysis, this is used to discover topics in a corpus using the bag-of-words document representation. Here we treat object categories as topics, so that an image containing instances of several categories is modeled as a mixture of topics. The model is applied to images by using a visual analogue of a word, formed by vector quantizing SIFT-like region descriptors. The topic discovery approach successfully translates to the visual domain: for a small set of objects, we show that both the object categories and their approximate spatial layout are found without supervision. Performance of this unsupervised method is compared to the supervised approach of Fergus et al. (2003) on a set of unseen images containing only one object per image. We also extend the bag-of-words vocabulary to include 'doublets' which encode spatially local co-occurring regions. It is demonstrated that this extended vocabulary gives a cleaner image segmentation. Finally, the classification and segmentation methods are applied to a set of images containing multiple objects per image. These results demonstrate that we can successfully build object class models from an unsupervised analysis of images. JF - Computer Vision, 2005. ICCV 2005. Tenth IEEE International Conference on EP - 377 TI - Discovering objects and their location in images VL - 1 T2 - 2005 IEEE International Conference on Computer Vision (ICCV) SP - 370 KW - education AU - Sivic, J AU - Russell, BC AU - Efros, AA AU - Zisserman, A AU - Freeman, WT PY - 2005/// UR - http://dx.doi.org/10.1109/ICCV.2005.77 ER -