Information fusion in content based image retrieval: A comprehensive overview

Piras, Luca; Giacinto, Giorgio

doi:10.1016/j.inffus.2017.01.003

Cited by 114 publications

(37 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We adopt the early fusion approach to combine several image descriptors [24], [25], by concatenate the feature vectors into a single vector. An image is represented as = ( ), where is a set of low level feature, namely color ( ) and shape ( ).…”

Section: ) Hybrid Featuresmentioning

confidence: 99%

Toward a New Approach in Fruit Recognition using Hybrid RGBD Features and Fruit Hierarchy Property

Rachmawati¹,

Supriana²,

Khodra³

2017

EECSI

View full text Add to dashboard Cite

We present hierarchical multi-feature classification (HMC) system for multiclass fruit recognition problem. Our approach to HMC exploits the advantages of combining multimodal features and the fruit hierarchy property. In the construction of hybrid features, we take the advantage of using color feature in the fruit recognition problem and combine it with 3D shape feature of depth channel of RGBD (Red, Green, Blue, Depth) images. Meanwhile, given a set of fruit species and variety, with a preexisting hierarchy among them, we consider the problem of assigning images to one of these fruit variety from the point of view of a hierarchy. We report on computational experiment using this approach. We show that the use of hierarchy structure along with hybrid RGBD features can improve the classification performance.

show abstract

Section: ) Hybrid Featuresmentioning

confidence: 99%

Toward a New Approach in Fruit Recognition using Hybrid RGBD Features and Fruit Hierarchy Property

Rachmawati¹,

Supriana²,

Khodra³

2017

EECSI

View full text Add to dashboard Cite

show abstract

“…The Content-Based Image Retrieval (CBIR) systems are originally based on the use of descriptors [5], [6] for encoding and retrieving images based on visual properties. In fact, a myriad of different descriptors are available, often providing distinct and complementary results even for a same query [7]. Different categories of descriptors may be more appropriate to certain problems than others.…”

Section: Introductionmentioning

confidence: 99%

“…Different categories of descriptors may be more appropriate to certain problems than others. A classic and simple example that represents this matter is when we compare the image of an orange to the one of a lemon [7]. By considering a shape descriptor, the score of similarity between them tends to be very high.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Selective Rank Fusion on Content-Based Image Retrieval

Valem

Pedronette

2019

Anais Estendidos Do XXXII Conference on Graphics, Patterns and Images (SIBRAPI Estendido 2019)

View full text Add to dashboard Cite

Mainly due to the evolution of technologies to store and share images, the growth of image collections have been remarkable for years. Therefore, developing effective methods to index and retrieve such extensive available visual information is indispensable. The CBIR (Content-Based Image Retrieval) systems are one of the main solutions for image retrieval tasks. These systems are mainly supported by the use of different visual descriptors and machine learning methods. Despite the relevant advances in the area, mainly driven by deep learning technologies, accurately computing the similarity between images remains a complex task in various scenarios due to the well known semantic gap problem. As distinct features produce complementary ranking results with different effectiveness performance, a promising solution consists in combining them. However, how to decide which visual features to combine is a very challenging task. This work proposes three novel methods for selecting and combining ranked lists by estimating their effectiveness in an unsupervised way. The approaches were evaluated in five different image collections and several descriptors, achieving results comparable or superior to the state-of-the-art in most of the evaluated scenarios.

show abstract

“…However, while integration can happen at different stages, efficiency and scalability of the integration scheme are also some of the most critical challenges in many application settings. Moreover, in most cases, multi-view integration schemes simply employ a linear model and, therefore, do not extensively utilize the relationships observed across different viewspecific representations [1], [2]. In fact, since multi-view feature distributions commonly vary quite significantly, such linear models often are not sufficiently expressive to obtain a comprehensive derived representation that can capture the complex association patterns with sufficient reliability.…”

Section: Introductionmentioning

confidence: 99%

Context-Aware Deep Sequence Learning with Multi-View Factor Pooling for Time Series Classification

Bhattacharjee

Tolone

Elshambakey

et al. 2018

2018 IEEE International Conference on Big Data (Big Data)

View full text Add to dashboard Cite

In this paper, we propose an effective, multiview, multivariate deep classification model for time-series data. Multi-view methods show promise in their ability to learn correlation and exclusivity properties across different independent information resources. However, most current multi-view integration schemes employ only a linear model and, therefore, do not extensively utilize the relationships observed across different view-specific representations. Moreover, the majority of these methods rely exclusively on sophisticated, handcrafted features to capture local data patterns and, thus, depend heavily on large collections of labeled data. The multiview, multivariate deep classification model for time-series data proposed in this paper makes important contributions to address these limitations. The proposed model derives a LSTM-based, deep feature descriptor to model both the viewspecific data characteristics and cross-view interaction in an integrated deep architecture while driving the learning phase in a data-driven manner. The proposed model employs a compact context descriptor to exploit view-specific affinity information to design a more insightful context representation. Finally, the model uses a multi-view factor-pooling scheme for a contextdriven attention learning strategy to weigh the most relevant feature dimensions while eliminating noise from the resulting fused descriptor. As shown by experiments, compared to the existing multi-view methods, the proposed multi-view deep sequential learning approach improves classification performance by roughly 4% in the UCI multi-view activity recognition dataset, while also showing significantly robust generalized representation capacity against its single-view counterparts, in classifying several large-scale multi-view light curve collections.

show abstract

Information fusion in content based image retrieval: A comprehensive overview

Cited by 114 publications

References 83 publications

Toward a New Approach in Fruit Recognition using Hybrid RGBD Features and Fruit Hierarchy Property

Toward a New Approach in Fruit Recognition using Hybrid RGBD Features and Fruit Hierarchy Property

Unsupervised Selective Rank Fusion on Content-Based Image Retrieval

Context-Aware Deep Sequence Learning with Multi-View Factor Pooling for Time Series Classification

Contact Info

Product

Resources

About