We present a novel transfer learning approach to cross-camera action recognition. Inspired by canonical correlation analysis (CCA), we first extract the spatio-temporal visual words from videos captured at different views, and derive a correlation subspace as a joint representation for different bag-of-words models at different views. Different from prior CCA-based approaches which simply train...

This paper defines a systematic approach to Opinion Mining (OM) on YouTube comments by (i) modeling classifiers for predicting the opinion polarity and the type of comment and (ii) proposing robust shallow syntactic structures for improving model adaptability. We rely on the tree kernel technology to automatically extract and learn features with better generalization power than bag-of-words. An...

We recently reported that patients who had received unilateral temporal lobectomy, including the amygdala and hippocampus, show impaired acquisition in a fear conditioning task (LaBar, LeDoux, Spencer, & Phelps, 1995), indicating a deficit in emotional memory. In the present paper, we examined performance of these patients on two verbal, emotional memory tasks in an effort to determine the exte...

introduction: effective perception of fear signals is crucial for organism survival. when threated, the organism indicates defensive behaviors. methods: elevated plus–maze has high efficiency for measurement of fear behavior and is widly used for fear behavior determination. increase in two parameters percent of openarm entries (%oae) and percent of time spent in the open-arm (%oat) in the elev...

Fear of crime is a very prevalent issue today. It challenges the safety of cities. For this reason it has gained considerable attention among the sociologists and criminologists over the last 20 years. Whilst men experiencing relatively high rates of violence, they report their fear of crime less than women. In the contrary women have a much greater fear of crime than men, but are victimized le...

Does SVM really scale up to large bag of words feature spaces? Motivation

This paper shows how semantic attributes can be used to improve object classification. The semantic attributes used fall into five groups: scene (e.g. ‘road’), colour (e.g. ‘green’), part (e.g. ‘face’), shape (e.g. ‘box’), and material (e.g. ‘wood’). We train a set of classifiers for individual semantic attributes, and use them to make predictions on new images (Figure 1). We can then use the s...