Image Retrieval with Fisher Vectors of Binary Features

Uchida, Yusuke; Sakazawa, Shigeyuki

doi:10.1109/acpr.2013.6

Cited by 18 publications

(37 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The existing frameworks do not use two descriptors that are extracted simultaneously for different purposes. Our framework is as robust as typical frameworks that use SIFT [1], [7], [8], and it can reduce both the time needed to extract local descriptors and storage size as well as an approach that uses binary descriptors [6].…”

Section: Introductionmentioning

confidence: 97%

“…The binary descriptors such as CARD, ORB, and BRISK obtain binary values by using simple binary tests between pixels in a smoothed image patch instead of computing gradients from the patch [3]- [5], [9]. [6]. Their approach computes Fisher vectors by aggregating ORB descriptors under the assumption that they are generated from a Bernoulli mixture model [12].…”

Section: Local Descriptor Extraction and Descriptor Aggregationmentioning

confidence: 99%

“…Moreover, a bag of visual words vector (BOVW) was calculated using SIFT in order to compare our approach with BOVW, which is a popular aggregation method [7]. In particular, ORB was converted into a Fisher vector by using the method described by Uchida et al [6].…”

Section: Comparative Experimentsmentioning

confidence: 99%

“…The binary descriptors such as ORB and BRISK that have bits at each keypoint [3]- [5]. Aggregation vectors with binary descriptors are not as robust as aggregation vectors with real-valued descriptors such as SIFT [6]. Thus, this approach sacrifices accuracy of image retrieval.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Image Retrieval Framework Based on Dual Representation Descriptor

Yoshida

Toyofuku

2017

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYDescriptor aggregation techniques such as the Fisher vector and vector of locally aggregated descriptors (VLAD) are used in most image retrieval frameworks. It takes some time to extract local descriptors, and the geometric verification requires storage if a real-valued descriptor such as SIFT is used. Moreover, if we apply binary descriptors to such a framework, the performance of image retrieval is not better than if we use a real-valued descriptor. Our approach tackles these issues by using a dual representation descriptor that has advantages of being both a real-valued and a binary descriptor. The real value of the dual representation descriptor is aggregated into a VLAD in order to achieve high accuracy in the image retrieval, and the binary one is used to find correspondences in the geometric verification stage in order to reduce the amount of storage needed. We implemented a dual representation descriptor extracted in semi-real time by using the CARD descriptor. We evaluated the accuracy of our image retrieval framework including the geometric verification on three datasets (holidays, ukbench and Stanford mobile visual search). The results indicate that our framework is as accurate as the framework that uses SIFT. In addition, the experiments show that the image retrieval speed and storage requirements of our framework are as efficient as those of a framework that uses ORB. key words: image retrieval, local descriptor

show abstract

Section: Introductionmentioning

confidence: 97%

Section: Local Descriptor Extraction and Descriptor Aggregationmentioning

confidence: 99%

Section: Comparative Experimentsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Image Retrieval Framework Based on Dual Representation Descriptor

Yoshida

Toyofuku

2017

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Using a more principled approach, Zhang et al [44] proposed a learning scheme based on the Hamming distance that proved to be useful in classification. More related to our work, Uchida and Sakazawa [42] derived a FV based on mixtures of Bernoulli pdfs which was shown to perform better than the BoBW in an object retrieval task. In [6], the authors propose a model that extends the BoVW by computing histograms of distances between the set of descriptors and each element in the codebook, learned using the k-medians algorithm and the Hamming distance.…”

Section: Introductionmentioning

confidence: 98%

Exponential family Fisher vector for image classification

Sánchez

Redolfi

2015

Pattern Recognition Letters

View full text Add to dashboard Cite

a b s t r a c tOne of the fundamental problems in image classification is to devise models that allow us to relate the images to higher-level semantic concepts in an efficient and reliable way. A widely used approach consists on extracting local descriptors from the images and to summarize them into an image-level representation. Within this framework, the Fisher vector (FV) is one of the most robust signatures to date. In the FV, local descriptors are modeled as samples drawn from a mixture of Gaussian pdfs. An image is represented by a gradient vector characterizing the distributions of samples w.r.t. the model. Equipped with robust features like SIFT, the FV has shown state-of-the-art performance on different recognition problems. However, it is not clear how it should be applied when the feature space is clearly non-Euclidean, leading to heuristics that ignore the underlying structure of the space. In this paper we generalize the Gaussian FV to a broader family of distributions known as the exponential family. The model, termed exponential family Fisher vectors (eFV), provides a unified framework from which rich and powerful representations can be derived. Experimental results show the generality and flexibility of our approach.

show abstract