Video data mining using configurations of viewpoint invariant regions

By Josef Sivic, Andrew Zisserman. 2004

204 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

SUN database: Large-scale scene recognition from abbey to zoo (2010, methodology): “Sparse SIFT histograms: As in “Video Google” [27], we build SIFT features at Hessian-affine and MSER [19] interest points.”
Fast image search for learned metrics (2008, background): “, bags of words or histogram pyramids [24, 13]).”
Weakly-supervised Discovery of Visual Pattern Configurations (2014, background): “Modeling the spatial and geometric relationship between co-occurring visual patterns (objects or object-parts) often improves visual recognition performance [30, 9, 25, 21, 31, 16, 8, 6, 22, 17].”
A Comparison of Affine Region Detectors (2005, unknown): “As such, even though they have often been called invariant regions in the literature (e.g., [5, 13, 41, 45]), in principle they should be termed covariant regions since they change covariantly with the transformation.”
Proceedings of the 5th International Conference on Image and Video Retrieval (2006, unknown): “without motion segmentation, akin to what proposed by [5].”
A survey on activity recognition and behavior understanding in video surveillance (2013, unknown): “in [58, 90, 134, 146, 176].”
Learning query-dependent prefilters for scalable image retrieval (2009, methodology): “A second approach is to mine a training set for frequently occurring word tuples, for example using “frequent itemset mining” [21], or by spatial selection of word tuples [20, 10, 23], or combinations of the two [22].”
World-scale mining of objects and events from community photo collections (2008, background): “Mining objects from visual data has been proposed for video for instance in [28, 21].”
Human vs. Computer in Scene and Object Recognition (2014, methodology): “Reference [20] [22] [8] [8] [15] [6] [15] [10] [15, 10] [4] [4] [9] [9] [2] [2] [7] [13] [12] [12] [11] [11] [14] Feat.”
Foreground Focus: Unsupervised Learning from Partially Matching Images (2009, background): “…Agarwal and Triggs (2006) neighborhoods are collected hierarchically in space, in Quack et al. (2007) a tiled region is centered on each interest point to bin nearby visual words, and in Sivic and Zisserman (2004) the k-nearest points to the base point are included but without any spatial ordering.”
Geometric min-Hashing: Finding a (thick) needle in a haystack (2009, methodology): “A similar property has been empirically observed and used in a different problem of correspondence matching [23] and object retrieval [25].”
Mining Mid-level Features for Image Classification (2014, methodology): “Mid-level features are constructed from LBOW (as e.g. used in Agarwal and Triggs 2008; Quack et al. 2007; Sivic and Zisserman 2004; Yuan et al. 2007) representing the distribution of local features in a semi-local neighbourhood.”