Robust Nonnegative Patch Alignment for Dimensionality Reduction

Adaptive Weighted Sparse Principal Component Analysis for Robust Unsupervised Feature Selection

Current unsupervised feature selection methods cannot well select the effective features from the corrupted data. To this end, we propose a robust unsupervised feature selection method under the robust principal component analysis (PCA) reconstruction criterion, which is named the adaptive weighted sparse PCA (AW-SPCA). In the proposed method, both the regularization term and the reconstruction error term are constrained by the l_2,1 -norm: the l_2,1 -norm regularization term plays a role in the feature selection, while the l_2,1 -norm reconstruction error term plays a role in the robust reconstruction. The proposed method is in a convex formulation, and the selected features by it can be used for robust reconstruction and clustering. Experimental results demonstrate that the proposed method can obtain better reconstruction and clustering performance, especially for the corrupted data.

Learning to Map Social Network Users by Unified Manifold Alignment on Hypergraph

Nowadays, a lot of people possess accounts on multiple online social networks, e.g., Facebook and Twitter. These networks are overlapped, but the correspondences between their users are not explicitly given. Mapping common users across these social networks will be beneficial for applications such as cross-network recommendation. In recent years, a lot of mapping algorithms have been proposed which exploited social and/or profile relations between users from different networks. However, there is still a lack of unified mapping framework which can well exploit high-order relational information in both social structures and profiles. In this paper, we propose a unified hypergraph learning framework named unified manifold alignment on hypergraph (UMAH) for this task. UMAH models social structures and user profile relations in a unified hypergraph where the relative weights of profile hyperedges are determined automatically. Given a set of training user correspondences, a common subspace is learned by preserving the hypergraph structure as well as the correspondence relations of labeled users. UMAH intrinsically performs semisupervised manifold alignment with profile information for calibration. For a target user in one network, UMAH ranks all the users in the other network by their probabilities of being the corresponding user (measured by similarity in the subspace). In experiments, we evaluate UMAH on three real world data sets and compare it to state-of-art baseline methods. Experimental results have demonstrated the effectiveness of UMAH in mapping users across networks.

Face recognition for video surveillance with aligned facial landmarks learning

BACKGROUND:

Video-based face recognition has attracted much attention owning to its wide range of applications such as video surveillance. There are various approaches for facial feature extraction. Feature vectors extracted by these approaches tend to have large dimension and may include redundant information for face representation, which limits the application of methods with high accuracy such as machine learning.

OBJECTIVE:

Facial landmarks represent the intrinsic characteristics of human face, which can be utilized to decrease redundant information and reduce the computation complexity. But feature points extracted in each frame of a video are irregular which needed to be aligned.

METHODS:

This paper presents a novel method which is based on facial landmarks and machine learning. We proposed a method to align the feature data into a common co-ordinate frame, and use a robust AdaBoost algorithm for classification.

RESULTS:

Experiments on the public Honda/UCSD database demonstrate the superior performance of our method to several state-of-the-art approaches. Experiments on Yale database show the sensitivity and specificity of the proposed method.

CONCLUSION:

The proposed methods can improve the image-set based recognition performance.