No abstract

No abstract

Conventional image categorization techniques primarily rely on low-level visual cues. In this paper, we describe a multimodal fusion scheme which improves the image classification accuracy by incorporating the information derived from the embedded texts detected in the image under classification. Specific to each image category, a text concept is first learned from a set of labeled texts in images of the target category using Multiple Instance Learning [1]. For an image under classification which contains multiple detected text lines, we calculate a weighted Euclidian distance between each text line and the learned text concept of the target category. Subsequently, the minimum distance, along with lowlevel visual cues, are jointly used as the features for SVM-based classification. Experiments on a challenging image database demonstrate that the proposed fusion framework achieves a higher accuracy than the state-of-art methods for image classification.

No abstract

Patients with osteoporosis are at significantly greater risk of developing SSNHL.

No abstract

No abstract

In this paper, we investigate the problem of automatically constructing characters' social network from movies. Unlike existing approaches that use co-appearance information to measure the relationship between two characters, we argue that a method that describes the characters' interaction, rather than the co-appearance, makes more sense. We propose a new scheme that quantifies the interaction of characters by the use of film-editing cues, based on which we construct the characters' social network. Experiments on real-world data validate the effectiveness of the proposed method. In addition, we show an application of discovering characters' social clusters enabled by the automatically constructed social network.