Modeling scenes with local descriptors and latent aspects

By Pedro Quelhas, Florent Monay, Jean-Marc Odobez, Daniel Gatica-Perez, Tinne Tuytelaars, Luc Van Gool. Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1, 2005

408 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/ba6417baed41a8f0fd4cab342aa214704389dcf9

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

In Media Technology

Citations in the corpus

Integrated visual vocabulary in latent Dirichlet allocation–based scene classification for IKONOS image (2014, background): “(3) is its denominator, which represents the summation over KW terms.”
Learning Object Categories From Internet Image Searches (2010, background): “Quelhas et al. [31] and Bosch et al. [5] perform scene classification.”
Supervised Topic Models (2007, background): “The hope was that LDA topics would turn out to be useful for categorization, since they act to reduce data dimension (Blei et al., 2003; Fei-Fei and Perona, 2005; Quelhas et al., 2005).”
Foreground Focus: Unsupervised Learning from Partially Matching Images (2009, background): “Early models transferred the notion of text documents containing unordered words to images composed of “visual words” (Quelhas et al. 2005; Fei-Fei and Perona 2005; Sivic et al. 2005).”
Linear spatial pyramid matching using sparse coding for image classification (2009, background): “Over the years many works have been done to improve the traditional BoF model, such as generative methods in [7, 21, 3, 1] for modeling the co-occurrence of the codewords or descriptors, discriminative codebook learning in [10, 5, 19, 27] instead of standard unsupervised K-means clustering, and spat”
Scene Classification Using a Hybrid Generative/Discriminative Approach (2008, background): “6, the gap between pLSA and BoW increases as the number of labeled training images decreases, as was demonstrated in [25].”
Which is the best way to organize/classify images by content? (2007, methodology): “…image segmentation by using the mean shift technique, (ii) binary image region classification by using the SVM classifiers with an automatic scheme for searching the optimal model parameters and (iii) label-based aggregation of the connected similar image regions for salient object generation.”
SIFT and color feature fusion using localized maximum-margin learning for scene classification (2011, background): “One of the most popular and successful models is called the bags of visual words [11-13], which has a number of variants [14, 15] [16, 17].”
Automatic landslide detection from remote-sensing imagery using a scene classification method based on BoVW and pLSA (2013, background): “…community has found the bag-of-visual-words (BoVW) representation to be an effective image representation for tasks varying from object recognition, scene classification, to image retrieval (Sivic and Zisserman 2003; Quelhas et al. 2005; Zhang et al. 2006; Yang et al. 2007; Zhang et al. 2007).”
Topic regression multi-modal Latent Dirichlet Allocation for image annotation (2010, methodology): “Inspired by the success of the recent work on scene modeling [7] [12], we borrow a tool from statistical text document analysis and adopt a bag-of-word representation for both image and text.”
Large Scale Online Learning of Image Similarity Through Ranking (2009, methodology): “The mapping of the descriptors to discrete indexes is performed according to a codebook C, which is typically learned from the local descriptors of the training images through k-means clustering (Duygulu et al., 2002; Jeon and Manmatha, 2004; Quelhas et al., 2005 ).”
Scene Classification Via pLSA (2006, methodology): “[12] who also use a combination of pLSA and supervised classification.”
Using Multiple Segmentations to Discover Objects and their Extent in Image Collections (2006, background): “One major issue noticed by several groups [17, 21], is that the “visual words” are not always as descriptive as their text counterparts.”
Aggregating Continuous Word Embeddings for Information Retrieval (2013, methodology): “As another example, pLSA, LDA and their many variations have been extensively applied to problems such as image classiﬁcation (Quelhas et al., 2005) or object discovery (Russell et al., 2006).”
LabelMe: A Database and Web-Based Tool for Image Annotation (2008, background): “Even algorithms requiring no supervision (Sivic et al. 2005; Russell et al. 2006; Fei-Fei et al. 2003; Sudderth et al. 2005b, 2005a; Quelhas et al. 2005) need this quantitative framework.”
Shape discovery from unlabeled image collections (2009, background): “The first group considers ways to discover latent visual topics using models developed for text, such as pLSA and LDA [23, 6, 24, 17].”