Discriminative Object Class Models of Appearance and Shape by Correlatons

By Silvio Savarese, John M. Winn, Antonio Criminisi. 2006

257 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

CrossTransformers: spatially-aware few-shot transfer (2020, background): “Discriminative parts [5, 10, 22, 34, 42, 83] and visual words [74, 100] have a rich history, and have found applications in deformable-parts models [26], and classification [70, 100].”
Nonparametric Scene Parsing via Label Transfer (2011, background): “However, these learning-based methods do not, in general, scale well with the number of object categories.”
Real-Time Hand Gesture Detection and Recognition Using Bag-of-Features and Support Vector Machine Techniques (2011, unknown): “There were a lot of researches on expanding “bag-of-features” to include the spatial relation in the context of object categorization [17], [49]–[52].”
Learning human actions via information maximization (2008, methodology): “In our work, we apply the correlogram which has been successively applied for image and scene classification [7, 17].”
Mining Mid-level Features for Image Classification (2014, methodology): “Finally, some methods (e.g. Chum et al. 2009; Ling and Soatto 2007; Liu et al. 2008; Savarese et al. 2006; Yang and Newsam 2011; Yimeng Zhang 2009) have looked at constructing mid-level features by combining multiple low-level features.”
Auto-Context and Its Application to High-Level Vision Tasks and 3D Brain Image Segmentation (2010, background): “From the point of view of using context information, there have been a lot of recent work proposed in object recognition and scene understanding [8, 19, 16, 22, 18, 28, 7, 20].”
Modeling spatial layout with fisher vectors for image categorization (2011, background): “Visual word “bigrams” are considered in [23], by forming a bag-of-word representation over spatially neighboring image regions.”
Auto-context and its application to high-level vision tasks (2008, background): “From the point of view of using context information, there have been a lot of recent work proposed in object recognition and scene understanding [8, 19, 16, 22, 18, 28, 7, 20].”
Pyramid of Spatial Relatons for Scene-Level Land Use Classification (2015, methodology): “features can be very important for image classification [6], [7], [21], [31], [43].”
Spatial pyramid co-occurrence for image classification (2011, background): “Saverese et al. [12] propose a model which instead characterizes the relative locations.”
Grouplet: A structured image representation for recognizing human and object interactions (2010, methodology): “In order to remedy BoW, some methods have been proposed to either encode longer range image statistics [25, 24 ] or explicitly model spatial relationships among image patches [9, 7, 20].”
Sparse coding-based correlaton model for land-use scene classification in high-resolution remote-sensing images (2016, methodology): “Suggested by Savarese et al.,20 the kernel size and the number of kernels in the correlaton and sparse correlaton models were given a value of 80 and 10, respectively.”
Discriminative Hierarchical Modeling of Spatio-temporally Composable Human Activities (2014, background): “Some researchers have extended image representations, such as correlatons [28] and spatial pyramids [21] to videos [20], and have applied them to the problem of simple human action categorization.”
Shape and Appearance Context Modeling (2007, methodology): “It is interesting to note that Algorithm 1 is only marginally sensitive to the choice of the partition P , which, as opposed to [19], here is allowed to be arbitrary.”
Intelligent multi-camera video surveillance: A review (2013, methodology): “There are also other features such as correlograms (Huang et al., 1997) and correlations (Savarese et al., 2006) to capture the co-occurrence of visual words over spatial kernels. propose an approach of utilizing local motion features for person re-identification.”
Land-Use Scene Classification in High-Resolution Remote Sensing Images Using Improved Correlatons (2015, methodology): “This model was motivated by earlier work using correlograms of quantized colors for indexing and classifying images [13].”