Multiview Spectral Embedding

By Tian Xia, Dacheng Tao, Tao Mei, Yongdong Zhang. 2010

513 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Ling Shao
3. Luc Van Gool
4. Ming-Hsuan Yang
5. Yi Yang
6. Xiaogang Wang
7. Alan L. Yuille
8. Xuelong Li

Citations in the corpus

• One-shot-learning gesture recognition using HOG-HOF features (2013, methodology): “With the availability of such technologies, conventional input devices, such as a keyboard or mouse, could be replaced in situations in which they are inconvenient in future.”
• A Survey on Multi-view Learning (2013, methodology): “multi-view embedding, we conduct the embedding for multiple features simultaneously while considering the consistency and complement of different views. For example, the multi-view spectral embedding (Xia et al., 2010) first builds a patch for a sample on each view, in which the arbitrary point and it”
• Binary Multi-View Clustering (2019, methodology): “Recently, multi-view clustering by exploiting heterogeneous features of data has attracted considerable attention [5], [6], [20].”
• On building ensembles of stacked denoising auto-encoding classifiers and their further improvement (2018, background): “The authors of [58] use spectral diversity [59] to train some DNNs whose outputs are aggregated, while partial training of learners with distorted data sets in the diversifi-cation applied in [60] .”
• Explore Efficient Local Features from RGB-D Data for One-Shot Learning Gesture Recognition (2016, methodology): “…et al. [11] extracted extended motion history image (Extended-MHI) [14] from RGB and depth videos respectively, and then applied multi-view spectral embedding (MSE) algorithm [15] to fuse these two Extended-MHI features, and finally used maximum correlation coefficient to achieve gesture recognition.”
• One shot learning gesture recognition from RGBD images (2012, methodology): “In the paper of [26], the value of d is acquired by cross validation.”
• Parameter-Free Auto-Weighted Multiple Graph Learning: A Framework for Multiview Clustering and Semi-Supervised Classification (2016, methodology): “Thus, this thought can be modeled as the following problem min F 2C,µ m X v=1 ⇣ µ (v) ⌘ T r ⇣ F T L (v) F ⌘ , s.t. m X v=1 µ (v) = 1, µ (v) 0, (6) which is introduced in [Xia et al., 2010; Cai et al., 2013; Li et al., 2015].”
• Learning to Rank Using User Clicks and Visual Features for Image Retrieval (2015, methodology): “Furthermore, we adopt multiview distance metric learning [31], [41] to combine these features and describe the similarity between images through”
• Neighborhood repulsed metric learning for kinship verification (2012, background): “Previous studies have shown that different feature descriptors could provide complementary information in characterizing facial information from different viewpoints [26, 27], and hence it is desirable to utilize multiple feature information for our kinship verification task.”
• Large-Scale Heterogeneous Feature Embedding (2019, methodology): “To analyze the effectiveness of FeatWalk and baselines, we apply their learned embedding representations H to perform classification, following the commonly-used way of validating feature embedding methods (Xia et al. 2010; Zhang et al. 2015b).”
• Tracking via Robust Multi-task Multi-view Joint Sparse Representation (2013, background): “1Regarding the term multi-view learning [25] [29], we follow the machine learning convention, in which views refer to different feature subsets used to represent particular characteristics of an object.”
• GMC: Graph-Based Multi-View Clustering (2020, background): “Our work is also related to multi-view spectral clustering [18], [19], [20], [21], [22], [23], [24], [25].”
• Ensemble manifold regularized sparse low-rank approximation for multiview feature embedding (2015, methodology): “Since this strategy completely ignores the diversity of multiple features, researchers have suggested to consider different views in different ways to achieve a better feature embedding, such as Distributed Spectral Embedding (DSE) [20] and Multiview Spectral Embedding (MSE) [21].”
• On Combining Multiple Features for Hyperspectral Remote Sensing Image Classification (2012, background): “erties of multiple features, different views definitely have different contributions to the final low-dimensional combination [42], [43].”