Guangdong Feng

Learn More
The problem of recovering data in multi-way arrays (i.e., tensors) arises in many fields such as image processing and computer vision, etc. In this paper, we present a novel method based on multi-linear n-rank and l_0 norm optimization for recovering a low-n-rank tensor with an unknown fraction of its elements being arbitrarily corrupted. In the new method,(More)
The problem of recovering data in multi-way arrays (i.e., tensors) arises in many fields such as image processing and computer vision, etc. In this paper, we present a novel method based on multi-linear n-rank and l0 norm optimization for recovering a low-n-rank tensor with an unknown fraction of its elements being arbitrarily corrupted. In the new method,(More)
The Safe following distance is closely associated with vehicle braking process. The braking process consists of three stages. The key problem in this research area is to quantitatively describe each stage. In this paper a framework is presented to model the safe following distance by analyzing the braking process. In this framework we focus on the third(More)
The different face regions have different degrees of importance for face recognition. In previous Hausdorff distance (HD) measures, points are treated as same importance, or weight different points that calculated from gray domain. In this paper, a new weighting function of HD based on the eigenface from edge domain, which reflects the discriminative(More)
  • 1