Name
Abstract | ||
|---|---|---|
Manifold ranking (MR), as a powerful semi-supervised learning algorithm, plays an important role to deal with the relevance feedback problem in content-based image retrieval (CBIR). However, conventional MR has two main drawbacks: 1) in many cases, it is prone to exploit "unreliable" unlabeled images when deployed in CBIR due to the semantic gap; 2) the performance of MR is quite sensitive to the scale parameter used for calculating the Laplacian matrix. In this work, a self-immunizing MR approach is presented to address the drawbacks. Concretely, we first propose an elastic kNN graph as well as its constructing algorithm to exploit unlabeled images "safely", and then develop a local scaling solution to calculate the Laplacian matrix adaptively. Extensive experiments on 10,000 Corel images show that the proposed algorithm is more effective than the state-of-the-art approaches. © Springer-Verlag 2013. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1007/978-3-642-37456-2_36 | PAKDD (2) |
Keywords | Field | DocType |
content-based image retrieval,elastic knn graph,local scaling,relevance feedback,self-immunizing manifold ranking | Data mining,Relevance feedback,Ranking SVM,Computer science,Image retrieval,Manifold alignment,Artificial intelligence,Laplacian matrix,Pattern recognition,Semantic gap,Exploit,Content-based image retrieval,Machine learning | Conference |
Volume | Issue | ISSN |
7819 LNAI | PART 2 | 16113349 |
Citations | PageRank | References |
4 | 0.39 | 13 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jun Wu | 1 | 125 | 15.66 |
| Yidong Li | 2 | 151 | 43.42 |
| Songhe Feng | 3 | 218 | 34.57 |
| Hong Shen | 4 | 499 | 52.98 |