A study of Bag-of-Visual-Words representations for handwritten keyword spotting

Aldavert, David; Rusiñol, Marçal; Toledo, Ricardo; Lladós, Josep

doi:10.1007/s10032-015-0245-z

Cited by 48 publications

(30 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In [7,48], profile features were combined with the shape based structural features for a partial matching scheme using dtw. Although these features Variable Length Profile+Moments [46] Profile+DFT [40] Slit HoG [79] Local Gradient Histogram [58] HMM SIFT [63] SIFT [1,2,67,70,86] BoWs SIFT [3] Fisher SIFT [4] PHOC Attributes Deep Learning [37] Neural Codes Deep Learning [35,74,75,83] PHOC Attributes Deep Learning [24] Levenshtein Embedding are fast to compute, it is susceptible to noise and common degradation present in documents.…”

Section: Classical Representationmentioning

confidence: 99%

See 1 more Smart Citation

HWNet v2: an efficient word image representation for handwritten documents

Krishnan

Jawahar

2019

IJDAR

View full text Add to dashboard Cite

We present a framework for learning an efficient holistic representation for handwritten word images. The proposed method uses a deep convolutional neural network with traditional classification loss. The major strengths of our work lie in: (i) the efficient usage of synthetic data to pre-train a deep network, (ii) an adapted version of the ResNet-34 architecture with the region of interest pooling (referred to as HWNet v2) which learns discriminative features for variable sized word images, and (iii) a realistic augmentation of training data with multiple scales and distortions which mimics the natural process of handwriting. We further investigate the process of transfer learning to reduce the domain gap between synthetic and real domain, and also analyze the invariances learned at different layers of the network using visualization techniques proposed in the literature.Our representation leads to a state-of-the-art word spotting performance on standard handwritten datasets and historical manuscripts in different languages with minimal representation size. On the challenging iam dataset, our method is first to report an mAP of around 0.90 for word spotting with a representation size of just 32 dimensions. Furthermore, we also present results on printed document datasets in English and Indic scripts which validates the generic nature of the proposed framework for learning word image representation.

show abstract

Section: Classical Representationmentioning

confidence: 99%

“…The popularity of bag of words (bow) [11,73] framework using local gradient features such as sift and hog, led to its proliferation to document images [1,2,67,68,70,86]. Rusinol et al [67,68], presented a patch based framework using bow histograms computed from the underlying sift descriptors.…”

Section: Bag Of Word Representationmentioning

confidence: 99%

HWNet v2: an efficient word image representation for handwritten documents

Krishnan

Jawahar

2019

IJDAR

View full text Add to dashboard Cite

show abstract

“…Thus there is no explicit use of visual word positions within the image. Traditional visual words based methods suffer when faced with similar appearances but distinct semantic concepts (Aldavert et al, 2015). In this study, we assume that establishing spatial dependencies might be useful for preserving the spatial organization of data.…”

Section: Relative Conjunction Matrixmentioning

confidence: 99%

Improving Bag-of-Visual-Words Towards Effective Facial Expressive Image Classification

Chanti

Caplier

2018

Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

Bag-of-Visual-Words (BoVW) approach has been widely used in the recent years for image classification purposes. However, the limitations regarding optimal feature selection, clustering technique, the lack of spatial organization of the data and the weighting of visual words are crucial. These factors affect the stability of the model and reduce performance. We propose to develop an algorithm based on BoVW for facial expression analysis which goes beyond those limitations. Thus the visual codebook is built by using k-Means++ method to avoid poor clustering. To exploit reliable low level features, we search for the best feature detector that avoids locating a large number of keypoints which do not contribute to the classification process. Then, we propose to compute the relative conjunction matrix in order to preserve the spatial order of the data by coding the relationships among visual words. In addition, a weighting scheme that reflects how important a visual word is with respect to a given image is introduced. We speed up the learning process by using histogram intersection kernel by Support Vector Machine to learn a discriminative classifier. The efficiency of the proposed algorithm is compared with standard bag of visual words method and with bag of visual words method with spatial pyramid. Extensive experiments on the CK+, the MMI and the JAFFE databases show good average recognition rates. Likewise, the ability to recognize spontaneous and non-basic expressive states is investigated using the DynEmo database.

show abstract

“…The Bag of word framework is among the popular and wellknown feature representation in information retrieval [8]. This method had been applied by J. Sivic and A. Zisserman in image and video retrieval field [9].…”

Section: Bag Of Word T Echniquesmentioning

confidence: 99%

A Performance Evaluation of Content-Based Image Retrieval for Scene Categorization

2017

Aug. 8-9, 2017 Singapore Back RTET-17, CFBHS-17, IECMAE-17, AESTWM-17, LHHSS-17 &Amp; LEBCSR-17

View full text Add to dashboard Cite

Abstract-Content-based image retrieval remains a critical problem in computer vision. In this paper, we study the performance of various content-based image retrieval technique for recognizing the object and scene. We conduct the comparative survey to compare the state of the art bag of words (BOW) framework with other method to help the researcher to understand more and enable researcher selecting the most suitable technique. We carried out the experiment and tested with 3 publicly dataset that are Caltech 101, Caltech 256 and 15-Scene Category dataset with BOW method. We also compare and evaluate the effect on the number of cluster toward the computational time and the accuracy. In addition, the significant of different feature extraction method ap plied in BOW performance is analyzed. In conclusion, we discuss on several key potential research topics towards the content-based image retrieval.

show abstract

A study of Bag-of-Visual-Words representations for handwritten keyword spotting

Cited by 48 publications

References 56 publications

HWNet v2: an efficient word image representation for handwritten documents

HWNet v2: an efficient word image representation for handwritten documents

Improving Bag-of-Visual-Words Towards Effective Facial Expressive Image Classification

A Performance Evaluation of Content-Based Image Retrieval for Scene Categorization

Contact Info

Product

Resources

About