LEARNING TO RANK IMAGE TAGS WITH LIMITED TRAINING EXAMPLES - IEEE PROJECTS IN PONDICHERRY,BULK IEEE PROJECTS IN PONDICHERRY,FINAL YEAR IEEE PROJECTS IN PONDICHERRY,BEST IEEE PROJECTS IN PONDICHERRY

LEARNING TO RANK IMAGE TAGS WITH LIMITED
TRAINING EXAMPLES
Abstract—With an increasing number of images that are available in social media,
image annotation has emerged as an important research topic due to its application
in image matching and retrieval. Most studies cast image annotation into a
multilabel classification problem. The main shortcoming of this approach is that it
requires a large number of training images with clean and complete annotations in
order to learn a reliable model for tag prediction. We address this limitation by
developing a novel approach that combines the strength of tag ranking with the
power of matrix recovery. Instead of having to make a binary decision for each tag,
our approach ranks tags in the descending order of their relevance to the given
image, significantly simplifying the problem. In addition, the proposed method
aggregates the prediction models for different tags into a matrix, and casts tag
ranking into a matrix recovery problem. It introduces the matrix trace norm to
explicitly control the model complexity, so that a reliable prediction model can be
learned for tag ranking even when the tag space is large and the number of training
images is limited. Experiments on multiple well-known image data sets
demonstrate the effectiveness of the proposed framework for tag ranking compared
with the state-of-the-art approaches for image annotation and tag ranking.

EXISTING SYSTEM:
In this section we review the related work on automatic image annotation and tag
ranking. Given the rich literature on both subjects, we only discuss the studies
closely related to this work, and refer the readers to for the detailed surveys of
these topics. A. Automatic Image Annotation Automatic image annotation aims to
find a subset of keywords/tags that describes the visual content of an image. It
plays an important role in bridging the semantic gap between low-level features
and high-level semantic content of images. Most automatic image annotation
algorithms can be classified into three categories (i) generative models that model
the joint distribution between tags and visual features, (ii) discriminative models
that view image annotation as a classification problem, and (iii) search based
approaches. Below, we will briefly review approaches in each category. Both
mixture models and topic models, two well known approaches in generative
model, have been successfully applied to automatic image annotation. a Gaussian
mixture model is used to model the dependence between keywords and visual
features. kernel density estimation is applied to model the distribution of visual
features and to estimate the conditional probability of keyword assignments given
the visual features. Topic models annotate images as samples from a specific
mixture of topics, which each topic is a joint distribution between image features

and annotation keywords. Various topic models have been developed for image
annotation, including probabilistic latent semantic analysis (pLSA) latent Dirichlet
allocation and hierarchical Dirichlet processes. Since a large number of training
examples are needed for estimating the joint probability distribution over both
features and keywords, the generative models are unable to handle the challenge of
large tag space with limited number of training images.
PROPOSED SYSTEM:
Although multiple algorithms have been developed for tag ranking, they tend to
perform poorly when the number of training images is limited compared to the
number of tags, a scenario often encountered in real world applications. In this
work, we address this limitation by casting tag ranking into a matrix recovery
problem. The key idea is to aggregate the prediction models for different tags into
a matrix. Instead of learning each prediction model independently, we propose to
learn all the prediction models simultaneously by exploring the theory of matrix
recovery, where a trace norm regularization is introduced to capture the
dependence among different tags and to control the model complexity. We shown,
both theoretically and empirically, that with the introduction of trace norm
regularizer, a reliable prediction model can be learned for tag ranking even when
the tag space is large and the number of training images is small. We note that

although the trace norm regularization has been studied extensively or
classificatio, this is the first study that exploits trace norm regularization for tag
ranking.
Module1
Automatic Image Annotation
Automatic image annotation aims to find a subset of keywords/tags that describes
the visual content of an image. It plays an important role in bridging the semantic
gap between low-level features and high-level semantic content of images. Most
automatic image annotation algorithms can be classified into three categories (i)
generative models that model the joint distribution between tags and visual
features, (ii) discriminative models that view image annotation as a classification
problem, and (iii) search based approaches. Below, we will briefly review
approaches in each category. Both mixture models and topic models, two well
known approaches in generative model, have been successfully applied to
automatic image annotation.A Gaussian mixture model is used to model the
dependence between keywords and visual features. kernel density estimation is
applied to model the distribution of visual features and to estimate the conditional
probability of keyword assignments given the visual features. Topic models
annotate images as samples from a specific mixture of topics, which each topic is a

joint distribution between image features and annotation keywords. Various topic
models have been developed for image annotation, including probabilistic latent
semantic analysis (pLSA) latent Dirichlet allocation and hierarchical Dirichlet
processes. Since a large number of training examples are needed for estimating the
joint probability distribution over both features and keywords, the generative
models are unable to handle the challenge of large tag space with limited number
of training images. A discriminative model views image annotation as a multi-
class classification problem, and learns one binary classification model for either
one or multiple tags. A 2D multiresolution hidded Markov model (MHMM) is
proposed to model the relationship between tags and visual content. A structured
max-margin algorithm is developed to exploit the dependence among tags. One
problem with discriminative approaches for image annotation is imbalanced data
distribution because each binary classifier is designed to distinguish image of one
class from images of the other classes. It becomes more severe when the number of
classes/tags is large. Another limitation of these approaches is that they are unable
to capture the correlation among classes, which is known to be important in multi-
label learning. To overcome these issues, several algorithms are proposed to
harness the keyword correlation as the additional information. The search based
approaches are based on the assumption that visually similar images are more
likely to share common keywords [10]. Given a test image I, it first finds out a set

of training images that are visually similar to I, and then assigns the tags that are
most popular among the similar images. A divide-and-conquer framework is
proposed which identifies the salient terms from textual descriptions of visual
neighbours searched from web images. In the Joint Equal Contribution (JEC)
model proposed, multiple distance functions are computed with each based on a
different set of visual features, and the nearest neighbors are determined by the
average distance functions. predicts keywords by taking a weighted combination of
tags assigned to nearest neighbor images. More recently, the sparse coding scheme
and its variations are employed to facilitate image label propagation. Similar to the
classification method, the search based approaches often fail when the number of
training examples is limited.
Module 2
Regularizedtag ranking
In this section, we first present the proposed framework for tag ranking that is
explicitly designed for a large tag space with a limited number of training images.
We then discuss a computational algorithm that efficiently solves the related
optimization problem. Regularization Framework for Tag Ranking In order to
learn a tag ranking function, we have to decide in the first place which tags are
relevant to a given image, and which ones are not. To his end, we simply assume

all the assigned tags are relevant, and the unassigned tags are irrelevant. Although
it is arguable that this simple treatment could be problematic for noisy and
incomplete tag assignments, it is justified by the empirical studywhere tag ranking
is shown to be more robust to both noisy and missing tags than the classification
approaches. As a result, we would like to learn a ranking function that assign a
higher score to tag t j than to a tag tk for image xi if y j i = 1 and yki = 0. More
specifically, let fi (x) be the prediction function or the i th tag, and let _(z) be a loss
function. Let ε j,k(x, y) measure the error in ranking tag t j and tk for image x with
respect to the true tag assignments y.
Module 3
Automatic Image Annotation and Tag Ranking
Given the learned matrix W∗ and a test image represented by vector xt , we
compute scores for different tags by yt = W_ ∗ xt that indicate the relevance of
each tag to the visual content of the test image. The tags are then ranked in the
descending order of the relevant scores and only the tags ranked at the top will be
used to annotate the test image. Besides image annotation, the learned model can
also be used when a subset of tags is provided to the test image and needs to be re-
ranked in order to remove the noisy tags.

CONCLUSION
In this work, we have proposed a novel tag ranking scheme for automatic image
annotation. The proposed scheme casts the tag ranking problem into a matrix
recovery problem and introduces trace norm regularization to control the model
complexity. Extensive experiments on image annotation and tag ranking have
demonstrated that the proposed method significantly outperforms several state-of-
the-art methods for image annotation especially when the number of training
images is limited and when many of the assigned image tags are missing. In the
future, we plan to apply the proposed framework to the image annotation problem
when image tags are acquired by crowdsouring that tend to be noisy and
incomplete.
REFERENCES
[1] R. Datta, D. Joshi, J. Li, and J. Z. Wang, “Image retrieval: Ideas, influences,
and trends of the new age,” ACM Comput. Surv., vol. 40, no. 2, 2008, Art. ID 5.
[2] J. Wu, H. Shen, Y. Li, Z.-B. Xiao, M.-Y. Lu, and C.-L. Wang, “Learning a
hybrid similarity measure for image retrieval,” Pattern Recognit., vol. 46, no. 11,
pp. 2927–2939, 2013.
[3] L. Wu, R. Jin, and A. K. Jain, “Tag completion for image retrieval,” IEEE
Trans. Pattern Anal. Mach. Intell., vol. 35, no. 3, pp. 716–727, Mar. 2013.

[4] A. Makadia, V. Pavlovic, and S. Kumar, “Baselines for image annotation,” Int.
J. Comput. Vis., vol. 90, no. 1, pp. 88–105, 2010.
[5] J. Tang, R. Hong, S. Yan, T.-S. Chua, G.-J. Qi, and R. Jain, “Image annotation
by kNN-sparse graph-based label propagation over noisily tagged web images,”
ACM Trans. Intell. Syst. Technol., vol. 2, no. 2, pp. 1–16, 2011.
[6] J. Tang, S. Yan, R. Hong, G.-J. Qi, and T.-S. Chua, “Inferring semantic
concepts from community-contributed images and noisy tags,” in Proc. 17th ACM
Int. Conf. Multimedia, 2009, pp. 223–232.
[7] M. Guillaumin, T. Mensink, J. Verbeek, and C. Schmid, “TagProp:
Discriminative metric learning in nearest neighbor models for image auto-
annotation,” in Proc. IEEE 12th Int. Conf. Comput. Vis., Sep./Oct. 2009, pp. 309–
316.
[8] W. Liu and D. Tao, “Multiview Hessian regularization for image annotation,”
IEEE Trans. ImageProcess., vol. 22, no. 7, pp. 2676–2687, Jul. 2013.
[9] S. Zhang, J. Huang, Y. Huang, Y. Yu, H. Li, and D. N. Metaxas, “Automatic
image annotation using group sparsity,” in Proc. IEEE Int. Conf. Comput. Vis.
Pattern Recognit., Jun. 2010, pp. 3312–3319.
[10] Y. Verma and C. V. Jawahar, “Image annotation using metric learning in
semantic neighbourhoods,” in Proc. 12th Eur. Conf. Comput. Vis., 2012, pp. 836–
849.

[11] Z. Feng, R. Jin, and A. Jain, “Large-scale image annotation by efficient and
robust kernel metric learning,” in Proc. IEEE Int. Conf. Comput. Vis., Dec. 2013,
pp. 1609–1616.
[12] G. Carneiro, A. B. Chan, P. J. Moreno, and N. Vasconcelos, “Supervised
learning of semantic classes for image annotation and retrieval,” IEEE Trans.
Pattern Anal. Mach. Intell., vol. 29, no. 3, pp. 394–410, Mar. 2007.

LEARNING TO RANK IMAGE TAGS WITH LIMITED TRAINING EXAMPLES - IEEE PROJECTS IN PONDICHERRY,BULK IEEE PROJECTS IN PONDICHERRY,FINAL YEAR IEEE PROJECTS IN PONDICHERRY,BEST IEEE PROJECTS IN PONDICHERRY

More Related Content

What's hot (10)

Similar to LEARNING TO RANK IMAGE TAGS WITH LIMITED TRAINING EXAMPLES - IEEE PROJECTS IN PONDICHERRY,BULK IEEE PROJECTS IN PONDICHERRY,FINAL YEAR IEEE PROJECTS IN PONDICHERRY,BEST IEEE PROJECTS IN PONDICHERRY (20)

More from Nexgen Technology (20)

Recently uploaded (20)

LEARNING TO RANK IMAGE TAGS WITH LIMITED TRAINING EXAMPLES - IEEE PROJECTS IN PONDICHERRY,BULK IEEE PROJECTS IN PONDICHERRY,FINAL YEAR IEEE PROJECTS IN PONDICHERRY,BEST IEEE PROJECTS IN PONDICHERRY