Block-Row Sparse Multiview Multilabel Learning for Image Classification
By Xiaofeng Zhu, Xuelong Li, Shichao Zhang. 2016
310 citations recorded by the source corpus.
In image analysis, the images are often represented by multiple visual features (also known as multiview features), that aim to better interpret them for achieving remarkable performance of the learning. Since the processes of feature extraction on each view are separated, the multiple visual features of images may include overlap, noise, and redundancy. Thus, learning with all the derived views of the data could decrease the effectiveness. To address this, this paper simultaneously conducts a hierarchical feature selection and a multiview multilabel (MVML) learning for multiview image classification, via embedding a proposed a new block-row regularizer into the MVML framework. The block-row regularizer concatenating a Frobenius norm (F-norm) regularizer and an l2,1-norm regularizer is designed to conduct a hierarchical feature selection, in which the F-norm regularizer is used to conduct a high-level feature selection for selecting the informative views (i.e., discarding the uninformative views) and the 12,1-norm regularizer is then used to conduct a low-level feature selection on the informative views. The rationale of the use of a block-row regularizer is to avoid the issue of the over-fitting (via the block-row regularizer), to remove redundant views and to preserve the natural group structures of data (via the F-norm regularizer), and to remove noisy features (the 12,1-norm regularizer), respectively. We further devise a computationally efficient algorithm to optimize the derived objective function and also theoretically prove the convergence of the proposed optimization method. Finally, the results on real image datasets show that the proposed method outperforms two baseline algorithms and three state-of-the-art algorithms in terms of classification performance.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Computer Vision and Pattern Recognition
Citations in the corpus
- A Survey on Multi-Task Learning (2017, background): “In [169], a multilabel multi-view algorithm is proposed to not only learn common features via the ‘2;1 regularization but also identify useless views via the Frobenius norm.”
- Learning k for kNN Classification (2017, methodology): “(3) The (cid:2) 1 -norm has been proved to lead the optimal sparse W ∗ [Zhu et al. 2016a; Wang et al. 2014; Zhang et al. 2011].”
- Deep Multimodal Distance Metric Learning Using Click Constraints for Image Ranking (2017, background): “Researchers have proposed some semi-supervised multiple kernel learning methods [24]–[26].”
- Efficient kNN classification algorithm for big data (2016, methodology): “Specifically, we propose to first conduct a k-means clustering to separate the whole dataset into several parts [26].”
- Robust Joint Graph Sparse Coding for Unsupervised Spectral Feature Selection (2017, background): “The mixed joint sparsity is the combination of the block sparsity and the row sparsity [54].”
- Feature selection for text classification: A review (2018, background): “…are some work targeted at novel applications of feature selection in recently years, for example, multi-label feature selection [42, 61, 82, 117, 118, 122], feature selection with streaming features [103], online feature selection [99], filter-based locality preserving feature selection [36],…”
- Efficient kNN Classification With Different Numbers of Nearest Neighbors (2018, background): “Equation (3) has been proved to result in sparse W [2], [56] and is also called the least absolute shrinkage and selection operator [9], [53].”