A Novel Saliency-Based Graph Learning Framework with Application to CBIR
スポンサーリンク
概要
- 論文の詳細を見る
Localized content-based image retrieval (LCBIR) has emerged as a hot topic more recently because in the scenario of CBIR, the user is interested in a portion of the image and the rest of the image is irrelevant. In this paper, we propose a novel region-level relevance feedback method to solve the LCBIR problem. Firstly, the visual attention model is employed to measure the regional saliency of each image in the feedback image set provided by the user. Secondly, the regions in the image set are constructed to form an affinity matrix and a novel propagation energy function is defined which takes both low-level visual features and regional significance into consideration. After the iteration, regions in the positive images with high confident scores are selected as the candidate query set to conduct the next-round retrieval task until the retrieval results are satisfactory. Experimental results conducted on the SIVAL dataset demonstrate the effectiveness of the proposed approach.
- （社）電子情報通信学会の論文
- 2011-06-01
著者
-
FENG Song-He
Institute of Computer & Engineering, Beijing Jiaotong University
-
Xu De
Institute Of Computer & Engineering Beijing Jiaotong University
-
Feng Song-he
Institute Of Computer & Engineering Beijing Jiaotong University
-
Liu Shuoyan
Institute Of Computer And Engineering Beijing Jiaotong University
-
BAO Hong
Institute of Computer & Engineering, Beijing Jiaotong University
-
Bao Hong
Institute Of Computer & Engineering Beijing Jiaotong University
-
Liu Shuoyan
Institute Of Computer & Engineering Beijing Jiaotong University
関連論文
- Color Constancy Based on Image Similarity
- How the Number of Interest Points Affect Scene Classification
- Natural Scene Classification Based on Integrated Topic Simplex
- Edge-Based Color Constancy via Support Vector Regression
- Combining Attention Model with Hierarchical Graph Representation for Region-Based Image Retrieval
- A Multi-Scale Adaptive Grey World Algorithm(Image Recognition, Computer Vision)
- A Novel Tone Mapping Based on Double-Anchoring Theory for Displaying HDR Images
- Adaptively Combining Local with Global Information for Natural Scenes Categorization
- Action Recognition Using Visual-Neuron Feature
- 2D Log-Gabor Wavelet Based Action Recognition
- Adaptive Non-linear Intensity Mapping Based Salient Region Extraction
- Modeling Bottom-Up Visual Attention for Color Images
- A Visual Inpainting Method Based on the Compressed Domain(Image Processing and Video Processing)
- Moving Object Completion on the Compressed Domain
- Color Constancy Based on Effective Regions
- Predicting DataSpace Retrieval Using Probabilistic Hidden Information
- Multi-Scale Multi-Level Generation Model in Scene Classification
- Discriminating Semantic Visual Words for Scene Classification
- A Novel Saliency-Based Graph Learning Framework with Application to CBIR
- Global-Context Based Salient Region Detection in Nature Images