Open AccessBook Chapter10.1007/978-3-319-16354-3_7

ConceptFusion: A Flexible Scene Classification Framework

- 29 Mar 2015

- pp 55-60

TL;DR: In this article, the authors proposed ConceptFusion, a method that combines the advantages of both low-level representations and high-level semantic categories, and eliminates the distinctions between different levels through the definition of concepts.

Abstract: We introduce ConceptFusion, a method that aims high accuracy in categorizing large number of scenes, while keeping the model relatively simpler and efficient for scalability The proposed method combines the advantages of both low-level representations and high-level semantic categories, and eliminates the distinctions between different levels through the definition of concepts The proposed framework encodes the perspectives brought through different concepts by considering them in concept groups that are ensembled for the final decision Experiments carried out on benchmark datasets show the effectiveness of incorporating concepts in different levels with different perspectives

AI Agents for this Paper

Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps

Most frequently asked questions

1. What have the authors contributed in "Conceptfusion: a flexible scene classification framework" ?

The authors introduce ConceptFusion, a method that aims high accuracy in categorizing large number of scenes, while keeping the model relatively simpler and efficient for scalability.

2. What are the future works mentioned in the paper "Conceptfusion: a flexible scene classification framework" ?

The authors plan to extend their framework by modifying this idea, and establishing dependence between each concept group by their semantic meanings.

Table 1. Comparison of different classifiers on MIT and SUN dataset

Table 2. Comparisons with feature concatenation and Object Bank [7] on MIT dataset

Fig. 1. Overview of ConceptFusion. Individual classifiers are trained for each concept group, and for each individual classifier a concept-priority value is computed. A test image, represented by concepts, is fed to the individual classifiers. Class-confidence values incorporated with the concept-priority values, are combined in the ensemble stage for final prediction.

Fig. 3. Comparing different number of concept groups on MIT (left) and SUN (right) datasets

Fig. 2. The effect of ensemble techniques in Sun Attribute (left) and MIT Indoor (right) datasets

References

•Journal Article•10.1023/A:1010933404324

Random Forests

Leo Breiman

- 01 Oct 2001

TL;DR: Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the forest, and are also applicable to regression.

...read moreread less

113.1K

Journal Article•10.1145/1961189.1961199

LIBSVM: A library for support vector machines

Chih-Chung Chang, +1 more

- 06 May 2011

- ACM Transactions on Intelligent Systems ...

TL;DR: Issues such as solving SVM optimization problems theoretical convergence multiclass classification probability estimates and parameter selection are discussed in detail.

...read moreread less

46.3K

•Proceedings Article•10.1109/CVPR.2005.177

Histograms of oriented gradients for human detection

Navneet Dalal, +1 more

- 20 Jun 2005

TL;DR: It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.

...read moreread less

36.7K

•Journal Article•10.1006/JCSS.1997.1504

A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting

Yoav Freund, +1 more

- 01 Aug 1997

TL;DR: The model studied can be interpreted as a broad, abstract extension of the well-studied on-line prediction model to a general decision-theoretic setting, and it is shown that the multiplicative weight-update Littlestone?Warmuth rule can be adapted to this model, yielding bounds that are slightly weaker in some cases, but applicable to a considerably more general class of learning problems.

...read moreread less

18.6K

•Proceedings Article•10.1109/CVPR.2006.68

Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories

Svetlana Lazebnik, +2 more

- 17 Jun 2006

TL;DR: This paper presents a method for recognizing scene categories based on approximate global geometric correspondence that exceeds the state of the art on the Caltech-101 database and achieves high accuracy on a large database of fifteen natural scene categories.

...read moreread less

9.2K

...

Expand

ConceptFusion: A Flexible Scene Classification Framework

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What have the authors contributed in "Conceptfusion: a flexible scene classification framework" ?

2. What are the future works mentioned in the paper "Conceptfusion: a flexible scene classification framework" ?

Figures

References

Random Forests

LIBSVM: A library for support vector machines

Histograms of oriented gradients for human detection

A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting

Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories

Related Papers (5)

Discriminative fields for modeling semantic concepts in video

Open Vocabulary Scene Parsing

Exploring Geometric Feature Hyper-Space in Data to Learn Representations of Abstract Concepts

Visual concepts and compositional voting

A context-aware semantic modeling framework for efficient image retrieval