Journal ArticleDOI
Unified Video Annotation via Multigraph Learning
TLDR
This paper shows that various crucial factors in video annotation, including multiple modalities, multiple distance functions, and temporal consistency, all correspond to different relationships among video units, and hence they can be represented by different graphs, and proposes optimized multigraph-based semi-supervised learning (OMG-SSL), which aims to simultaneously tackle these difficulties in a unified scheme.Abstract:
Learning-based video annotation is a promising approach to facilitating video retrieval and it can avoid the intensive labor costs of pure manual annotation. But it frequently encounters several difficulties, such as insufficiency of training data and the curse of dimensionality. In this paper, we propose a method named optimized multigraph-based semi-supervised learning (OMG-SSL), which aims to simultaneously tackle these difficulties in a unified scheme. We show that various crucial factors in video annotation, including multiple modalities, multiple distance functions, and temporal consistency, all correspond to different relationships among video units, and hence they can be represented by different graphs. Therefore, these factors can be simultaneously dealt with by learning with multiple graphs, namely, the proposed OMG-SSL approach. Different from the existing graph-based semi-supervised learning methods that only utilize one graph, OMG-SSL integrates multiple graphs into a regularization framework in order to sufficiently explore their complementation. We show that this scheme is equivalent to first fusing multiple graphs and then conducting semi-supervised learning on the fused graph. Through an optimization approach, it is able to assign suitable weights to the graphs. Furthermore, we show that the proposed method can be implemented through a computationally efficient iterative process. Extensive experiments on the TREC video retrieval evaluation (TRECVID) benchmark have demonstrated the effectiveness and efficiency of our proposed approach.read more
Citations
More filters
Journal Article
When is nearest neighbor meaningful
TL;DR: In this article, the authors explore the effect of dimensionality on the nearest neighbor problem and show that under a broad set of conditions (much broader than independent and identically distributed dimensions), as dimensionality increases, the distance to the nearest data point approaches the distance of the farthest data point.
Journal ArticleDOI
Recent advances and trends in visual tracking: A review
TL;DR: The goal of this paper is to review the state-of-the-art progress on visual tracking methods, classify them into different categories, as well as identify future trends.
Journal ArticleDOI
3-D Object Retrieval and Recognition With Hypergraph Analysis
TL;DR: A hypergraph analysis approach to address the problem of view-based 3-D object retrieval and recognition by avoiding the estimation of the distance between objects by constructing multiple hypergraphs based on their 2-D views.
Journal ArticleDOI
Click Prediction for Web Image Reranking Using Multimodal Sparse Coding
Jun Yu,Yong Rui,Dacheng Tao +2 more
TL;DR: A multimodal hypergraph learning-based sparse coding method is proposed for image click prediction, and the obtained click data is applied to the reranking of images, which shows the use of click prediction is beneficial to improving the performance of prominent graph-based image reranking algorithms.
Proceedings Article
Unsupervised feature selection using nonnegative spectral analysis
TL;DR: A new unsupervised learning algorithm, namely Nonnegative Discriminative Feature Selection (NDFS), which exploits the discriminative information and feature correlation simultaneously to select a better feature subset.
References
More filters
Proceedings Article
Distance Metric Learning for Large Margin Nearest Neighbor Classification
TL;DR: In this article, a Mahanalobis distance metric for k-NN classification is trained with the goal that the k-nearest neighbors always belong to the same class while examples from different classes are separated by a large margin.
Proceedings Article
Learning with Local and Global Consistency
TL;DR: A principled approach to semi-supervised learning is to design a classifying function which is sufficiently smooth with respect to the intrinsic structure collectively revealed by known labeled and unlabeled points.
BookDOI
Semi-Supervised Learning
TL;DR: Semi-supervised learning (SSL) as discussed by the authors is the middle ground between supervised learning (in which all training examples are labeled) and unsupervised training (where no label data are given).
Book ChapterDOI
When Is ''Nearest Neighbor'' Meaningful?
TL;DR: The effect of dimensionality on the "nearest neighbor" problem is explored, and it is shown that under a broad set of conditions, as dimensionality increases, the Distance to the nearest data point approaches the distance to the farthest data point.