Model Fusion in Conceptual Language Modeling

Maisonnasse, Loïc; Gaussier, Éric; Chevallet, Jean–Pierre

doi:10.1007/978-3-642-00958-7_23

Cited by 10 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the language model defined over the graph proposed in [15], we present in this paper an extension that handles set of concept sets and set of relation sets. The probability for a query image graph G Iq = < S Iq WC f , S Iq WE > to be generated from one document image graph G Id can be written as:…”

Section: Language Model For Graph Matchingmentioning

confidence: 99%

“…where #(c, Iq) denotes the number of times concept c occurs in the image query graph. This contribution corresponds to the concept probability as proposed in [15]. Similar to the previous work, the quantity P(c|G Id ) is estimated through maximum likelihood using Jelinek-Mercer smoothing:…”

Section: Concept Set Generationmentioning

confidence: 99%

“…This graph model is a generalization of the model defined in [21] which corresponds to the special case where only one concept set and one relation set are used. In some special cases, our model corresponds to the standard language model (LM) used in [15,20] where relations are not considered (i.e., documents and queries correspond to multiple bag-of-words model) .…”

Section: Relation Set Generationmentioning

confidence: 99%

See 2 more Smart Citations

Visual graph modeling for scene recognition and mobile robot localization

Pham

Mulhem

Maisonnasse

et al. 2010

Multimed Tools Appl

Self Cite

View full text Add to dashboard Cite

From the issue entitled "Content-based Multimedia Indexing"International audienceImage retrieval and categorization may need to consider several types of visual features and spatial information between them (e.g., different point of views of an image). This paper presents a novel approach that exploits an extension of the language modeling approach from information retrieval to the problem of graph-based image retrieval and categorization. Such versatile graph model is needed to represent the multiple points of views of images. A language model is defined on such graphs to handle a fast graph matching. We present the experiments achieved with several instances of the proposed model on two collections of images: one composed of 3,849 touristic images and another composed of 3,633 images captured by a mobile robot. Experimental results show that using visual graph model (VGM) improves the accuracies of the results of the standard language model (LM) and outperforms the Support Vector Machine (SVM) method

show abstract

Section: Language Model For Graph Matchingmentioning

confidence: 99%

Section: Concept Set Generationmentioning

confidence: 99%

Section: Relation Set Generationmentioning

confidence: 99%

See 1 more Smart Citation

Visual graph modeling for scene recognition and mobile robot localization

Pham

Mulhem

Maisonnasse

et al. 2010

Multimed Tools Appl

Self Cite

View full text Add to dashboard Cite

show abstract

“…The definitions of the functions F d and F D in the above equations are similar to the ones seen previously, but concern labels. The model we have just presented is inspired by the model defined in [6]. It differs however from it in (a) we propose in this paper a complete methodology for automatically indexing images at a conceptual level, and (b) it takes into account weights on each concept and association.…”

Section: A Language Model For Graph Matchingmentioning

confidence: 99%

“…Thus, Gao et al [2] proposes (a) the use of a dependency parser to represent documents and queries, and (b) an extension of the language modeling approach to deal with such trees. Maisonnasse [6] further extend this approach with a compatible model for general graphs, as the ones obtained by a conceptual analysis of documents and queries. We rely here on this latter models, however extending it by (a) applying it to an image collections, and (b) considering that both concepts and relations can be weighted.…”

Section: Introductionmentioning

confidence: 99%

Integration of spatial relationships in visual language model for scene retrieval

Pham

Mulhem

Maisonnasse

et al. 2010

2010 International Workshop on Content Based Multimedia Indexing (CBMI)

Self Cite

View full text Add to dashboard Cite

In this paper, we describe a method to use a graph-based language modeling approach for image retrieval and image categorization. We first mapped image regions to induced concepts and then spatial relationships between these regions to build a graph representation of images. Our method allows to deal with different scenarii, where isolated images or groups of images are used for training and testing. The results obtained on an image categorization problem comprising of 3849 images from 101 landmarks of Singapore show that (a) the procedure to automatically induce concepts from an image is effective, and (b) the use of spatial relationships, in addition to concepts, for representing an image content helps improve the classifier accuracy.This approach is the first one, to our knowledge, to present a complete extension of the language modeling approach from information retrieval to the problem of graph-based image categorization and retrieval.

show abstract

An Efficient Image Retrieval Using Medical-Dependent Features

Ayadi

Torjmen–Khemamkhem

2020

Digital Health in Focus of Predictive, Preventive and Personalised Medicine

View full text Add to dashboard Cite

Model Fusion in Conceptual Language Modeling

Cited by 10 publications

References 8 publications

Visual graph modeling for scene recognition and mobile robot localization

Visual graph modeling for scene recognition and mobile robot localization

Integration of spatial relationships in visual language model for scene retrieval

An Efficient Image Retrieval Using Medical-Dependent Features

Contact Info

Product

Resources

About