Query Specific Fusion for Image Retrieval

Zhang, Shaoting; Yang, Ming; Cour, Timothée; Yu, Kai; Metaxas, Dimitris N.

doi:10.1007/978-3-642-33709-3_47

Cited by 137 publications

(145 citation statements)

References 25 publications

Supporting

Mentioning

144

Contrasting

Order By: Relevance

“…[97] build an offline graph for each type of feature, which is subsequently fused during the online query. In an improvement of [97], Deng et al [98] add weakly supervised anchors to aid graph fusion. Both works on the rank level.…”

Section: Feature Fusionmentioning

confidence: 99%

SIFT Meets CNN: A Decade Survey of Instance Retrieval

Zheng

Yang

Tian

2018

IEEE Trans. Pattern Anal. Mach. Intell.

616

319

View full text Add to dashboard Cite

Abstract-In the early days, content-based image retrieval (CBIR) was studied with global features. Since 2003, image retrieval based on local descriptors (de facto SIFT) has been extensively studied for over a decade due to the advantage of SIFT in dealing with image transformations. Recently, image representations based on the convolutional neural network (CNN) have attracted increasing interest in the community and demonstrated impressive performance. Given this time of rapid evolution, this article provides a comprehensive survey of instance retrieval over the last decade. Two broad categories, SIFT-based and CNN-based methods, are presented. For the former, according to the codebook size, we organize the literature into using large/medium-sized/small codebooks. For the latter, we discuss three lines of methods, i.e., using pre-trained or fine-tuned CNN models, and hybrid methods. The first two perform a single-pass of an image to the network, while the last category employs a patch-based feature extraction scheme. This survey presents milestones in modern instance retrieval, reviews a broad selection of previous works in different categories, and provides insights on the connection between SIFT and CNN-based methods. After analyzing and comparing retrieval performance of different categories on several datasets, we discuss promising directions towards generic and specialized instance retrieval.

show abstract

Section: Feature Fusionmentioning

confidence: 99%

SIFT Meets CNN: A Decade Survey of Instance Retrieval

Zheng

Yang

Tian

2018

IEEE Trans. Pattern Anal. Mach. Intell.

616

319

View full text Add to dashboard Cite

show abstract

“…Therefore, RGB color space is not consistent with the perception of color psychology. But HSV color space is a kind of color model of visual perception ,the detail of this color model is analyzed in the reference paper [8],which is much closer to the people's experience and perception of color than RGB. It mainly includes three elements: Hue(H), Saturation (S),Value(V).…”

Section: A Color Space Selection and Transformationmentioning

confidence: 99%

“…Certainly, many researchers have found that image retrieval can get better result using two or more features of image [6][7][8][9].…”

Section: Introductionmentioning

confidence: 99%

Design and Implementation of Image Retrieval System Based on Block Color Moments and GLCM

Wang¹,

Ge²,

Xu³

et al. 2016

Proceedings of the 2016 International Conference on Computer Engineering and Information Systems

View full text Add to dashboard Cite

Abstract-In this paper, aiming to overcome the shortcoming of global color histogram which lack of color spatial distribution information, block color moment is used to describe the color feature. In order to extract the texture feature, Canny operator for edge detection is applied firstly, and then the gray level co-occurrence matrix of the edge image is constructed. The results prove that comprehensive of color and texture in image retrieval method overcomes the limitations of using a single, at the same time it improves the retrieval accuracy and speed.

show abstract

“…Multiple query methods and textual queries In [27] a query specific feature fusion method is proposed. Even though it is not presented as a MQIR method it can be extended to perform MQIR.…”

Section: Related Workmentioning

confidence: 99%