Localized content based image retrieval

Rahmani, Rouhollah; Goldman, Sally A.; Zhang, Hui; Krettek, J. E.; Fritts, Jason E.

doi:10.1145/1101826.1101863

Cited by 77 publications

(37 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We conduct compared experiments for domain adaptation on two benchmark datasets, 4 i.e., 20Newsgrous and In addition, two commonly used MIL datasets, i.e., Birds [56] and SIVAL [57] are also being used in experiments.…”

Section: Domain Adaptationmentioning

confidence: 99%

Towards Safe Weakly Supervised Learning

Guo

Zhou

2019

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

In this paper, we study weakly supervised learning where a large amount of data supervision is not accessible. This includes 4 i) incomplete supervision, where only a small subset of labels is given, such as semi-supervised learning and domain adaptation; ii) 5 inexact supervision, where only coarse-grained labels are given, such as multi-instance learning and iii) inaccurate supervision, where 6 the given labels are not always ground-truth, such as label noise learning. Unlike supervised learning which typically achieves 7 performance improvement with more labeled examples, weakly supervised learning may sometimes even degenerate performance 8 with more weakly supervised data. Such deficiency seriously hinders the deployment of weakly supervised learning to real tasks. It is 9 thus highly desired to study safe weakly supervised learning, which never seriously hurts performance. To this end, we present a 10 generic ensemble learning scheme to derive a safe prediction by integrating multiple weakly supervised learners. We optimize the 11 worst-case performance gain and lead to a maximin optimization. This brings multiple advantages to safe weakly supervised learning. 12 First, for many commonly used convex loss functions in classification and regression, it is guaranteed to derive a safe prediction under 13 a mild condition. Second, prior knowledge related to the weight of the base weakly supervised learners can be flexibly embedded. 14 Third, it can be globally and efficiently addressed by simple convex quadratic or linear program. Finally, it is in an intuitive geometric 15 interpretation with the least square loss. Extensive experiments on various weakly supervised learning tasks, including semi-16 supervised learning, domain adaptation, multi-instance learning and label noise learning demonstrate our effectiveness.

show abstract

Section: Domain Adaptationmentioning

confidence: 99%

Towards Safe Weakly Supervised Learning

Guo

Zhou

2019

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

show abstract

“…We have evaluated this proposed methods on three widely used data sets, the CIFAR-10 [29], the SIVAL (Spatially Independent, Variably Area, and Lighting) dataset [30] and the Oxford buildings dataset [31]. Each dataset is split into a training set and a query set.…”

Section: Data Setsmentioning

confidence: 99%

Spatial pyramid deep hashing for large-scale image retrieval

et al. 2017

View full text Add to dashboard Cite

Effective feature representations and similarity measurements are crucial for large-scale image retrieval, and conventional methods often learn hash functions from a predefined hand-crafted feature space. Meanwhile, the spatial structure in raw images always lost in most previous methods. Encouraged by the recent advances in convolutional neural networks(CNNs), a novel Spatial Pyramid Deep Hashing(SPDH) algorithm is developed for the task of fast image retrieval. In our SPDH algorithm, the CNN with a spatial pyramid pooling and a locally-connected layer with binary activation functions is utilized to build the end-to-end relation between the raw image data and the binary hashing codes for fast indexing. Different from the fully-connected layer, the locally-connected layer can consider each local spatial bin as an independent unit and only connect the local bin to preserve the spatial pyramid structure for hash codes. The learning of both the hash function and the feature representations are jointly optimized via backward propagation with classification or similarity loss function on the large-scale labeled dataset such as Ima-geNet. Moreover, a spatial pyramid binary pattern matching algorithm is developed to achieve partial local similar matching among the images. Our experimental results have shown that our SPDH method can outperform several state-of-the-art hashing algorithms on the CIFAR-10, SIVAL and the Oxford buildings datasets.

show abstract

“…The dedicated and special class of CBIR system that uses Blood-cell analysis method is described in [14]. The localized system is considered for the retrieval of the relevant portions of the image and demonstrated exclusively [15], whereas the relevant feedback based CBIR system is presented in [16,17].…”

Section: Relevant Workmentioning

confidence: 99%

Optimum Features selection by fusion using Genetic Algorithm in CBIR

Patil¹,

Chatur²,

Chaudhari³

2014

IJIGSP

View full text Add to dashboard Cite

The evaluation of the performance of the Content Based Image Retrieval is undertaken for the consideration in this paper. Here the point of the discussion is the performance of the CBIR system using object oriented image segmentation and the evolutionary computational technique. The visual characteristics of the objects such as color, intensity and texture are extracted by the conventional methods. Object oriented image segmentation along with the evolutionary computational technique is proposed here for Image Retrieval Algorithm. Unsupervised Curve evolution method is used for object oriented segmentation of the Image and genetic Algorithm is used for the Optimum Classification and reduction in the Feature dimensionality. The Algorithm is tested on the images which are characterized by the low depth. The Berkeley database is found to be suitable for this purpose. The experimental result shows that the Genetic Algorithm enhances the performance of this Content Based Image Retrieval and found to be suitable for optimization of features selection and compression technique for Feature space. Index terms-CBIR system, color histogram, edge histogram descriptor. Feature vector, genetic algorithm, Image retrieval, texture feature, query image. Dr. Prashant N Chatur has received his M.E. degree in Electronics Engineering from Govt. College of Engineering Amravati, India and Ph. D degree from Amravati University. His area of research includes ANN, Data Mining, Data Stream Mining and Cloud Computing. Currently he is Head of

show abstract

Localized content based image retrieval

Cited by 77 publications

References 12 publications

Towards Safe Weakly Supervised Learning

Towards Safe Weakly Supervised Learning

Spatial pyramid deep hashing for large-scale image retrieval

Optimum Features selection by fusion using Genetic Algorithm in CBIR

Contact Info

Product

Resources

About