Comparing discriminating transformations and SVM for learning during multimedia retrieval

Zhou, Xiang; Huang, Thomas S.

doi:10.1145/500141.500163

Cited by 75 publications

(48 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that the use of lower-dimensional features can also diminish the cost of kernel computations and the number of support vectors, thus contributing to a reduction in the cost of the selection stage (which nevertheless increases linearly with the size of the database). Motivated by the early analysis in (Zhou and Huang, 2001) highlighting the different nature of "relevant" and "irrelevant" feedback, a differentiated treatment of the items marked as "irrelevant" was proposed in (Tao et al, 2007) based on the kernel biased marginal convex machine. An alternative solution, introducing orthogonal complement component analysis, is suggested in (Tao et al, 2008).…”

Section: Relevance Feedback With Support Vector Machinesmentioning

confidence: 99%

Speeding up active relevance feedback with approximate kNN retrieval for hyperplane queries

Crucianu

Estevez

Oria

et al. 2008

Int J Imaging Syst Tech

View full text Add to dashboard Cite

In content-based image retrieval, relevance feedback is a prominent method for reducing the "semantic gap" between the low-level features describing the content and the usually higher-level meaning of user's target. Recent relevance feedback methods are able to identify complex target classes after relatively few feedback iterations. However, since the computational complexity of such methods is linear in the size of the database, retrieval can be quite slow on very large databases. To address this scalability issue for active learning-based relevance feedback, we put forward a method that consists in the construction of an index in the feature space associated to a kernel function and in performing approximate kNN hyperplane queries with this feature space index. The experimental evaluation performed on two image databases show that a significant speedup can be achieved, at the expense of a limited increase in the number of feedback rounds.

show abstract

Section: Relevance Feedback With Support Vector Machinesmentioning

confidence: 99%

Speeding up active relevance feedback with approximate kNN retrieval for hyperplane queries

Crucianu

Estevez

Oria

et al. 2008

Int J Imaging Syst Tech

View full text Add to dashboard Cite

show abstract

“…A further improvement has been made in [7], which shows that despite the importance of the NE in refining the results of image retrieval, the two models cited above do not support it, and then proposes a model that combines PE and NE in a single optimization problem. In [8], Zhou et al propose an RF model based on biased discriminant analysis (BDA), which combines PE and NE. In [32], Tao and Tang use nonparametric discriminant analysis (NDA) in order to avoid the parameter tuning problem and the single Gaussian distribution assumption in BDA.…”

Section: A Overview Of the State Of The Artmentioning

confidence: 99%

“…These include QBIC [1], Chabot [2], Simplicity [3], Mars [4], PicHunter [5], Virage [6], Atlas [7], and the system developed by Zhou et al [8]. Content-based image retrieval generally succeeds in meeting the needs of users interested in image visual content.…”

mentioning

confidence: 99%

Relevance feedback for CBIR: a new approach based on probabilistic feature weighting with positive and negative examples

Kherfi

Ziou

2006

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Abstract-In content-based image retrieval, understanding the user's needs is a challenging task that requires integrating him in the process of retrieval. Relevance feedback (RF) has proven to be an effective tool for taking the user's judgement into account. In this paper, we present a new RF framework based on a feature selection algorithm that nicely combines the advantages of a probabilistic formulation with those of using both the positive example (PE) and the negative example (NE). Through interaction with the user, our algorithm learns the importance he assigns to image features, and then applies the results obtained to define similarity measures that correspond better to his judgement. The use of the NE allows images undesired by the user to be discarded, thereby improving retrieval accuracy. As for the probabilistic formulation of the problem, it presents a multitude of advantages and opens the door to more modeling possibilities that achieve a good feature selection. It makes it possible to cluster the query data into classes, choose the probability law that best models each class, model missing data, and support queries with multiple PE and/or NE classes. The basic principle of our algorithm is to assign more importance to features with a high likelihood and those which distinguish well between PE classes and NE classes. The proposed algorithm was validated separately and in image retrieval context, and the experiments show that it performs a good feature selection and contributes to improving retrieval effectiveness.Index Terms-Content-based image retrieval (CBIR), feature selection (FS), relevance feedback (RF).

show abstract

“…For the computation of the distance, we used Biased Discriminant Analysis (BDA.) The detail of BDA is described in [33].…”

Section: Relevance Feedback Algorithmmentioning

confidence: 99%

Group-based interface for content-based image retrieval

Nakazato

Manola

Huang

2002

Proceedings of the Working Conference on Advanced Visual Interfaces

Self Cite

View full text Add to dashboard Cite

In Content-based Image Retrieval (CBIR) systems, the Query-byExample (QBE) approach is commonly used. However, because of inevitable "semantic gaps" between visual features and the user's concepts, trial-and-error query is essential for successful retrieval. Unfortunately, traditional user interfaces are not suitable for trying different combinations of query examples. This is because in these systems, query specification and result display are done on the same workspace. Once the user removes an image from the query examples, the image may disappear from the user interface. In addition, it is difficult to combine the result of different queries.In this paper, we propose a new interface for Content-based image retrieval. In our system, the users can interactively compare different combinations of query examples by dragging and grouping images on the workspace (Query-by-Group.) Because the query results are displayed on another pane, the user can quickly review the results. Combining different queries is also easy. Furthermore, the concept of "image groups" is also applied to annotating and organizing a large number of images. Because the gestural operations of our system is similar to file operations of modern window-based operation systems, users can easily learn to use the system.

show abstract

Comparing discriminating transformations and SVM for learning during multimedia retrieval

Cited by 75 publications

References 30 publications

Speeding up active relevance feedback with approximate kNN retrieval for hyperplane queries

Speeding up active relevance feedback with approximate kNN retrieval for hyperplane queries

Relevance feedback for CBIR: a new approach based on probabilistic feature weighting with positive and negative examples

Group-based interface for content-based image retrieval

Contact Info

Product

Resources

About