Multi-level feature representations for video semantic concept detection

详细信息    查看全文

• 作者：Haojie Li^a ; ^{hjli@dlut.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Lijuan Liu^a ; Fuming Sun^b ; ^{sunfm@mail.ustc.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Yu Bao^a ; Chenxin Liu^a
• 关键词：Video retrieval ; Concept detection ; Semantic gap ; Fisher Vector ; LDA
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：8 January 2016
• 年：2016
• 卷：172
• 期：Complete
• 页码：64-70
• 全文大小：1127 K

文摘

Video semantic concept detection is a fundamental problem with many practical applications such as concept-based video retrieval. The major challenge of concept detection lies in the existence of the well-known semantic gap between the low-level visual features and the user壮s semantic interpretation of visual data. To bridge the semantic gap, in this paper we propose to promote low-level visual features to middle-level representations, expecting that the underlying latent semantic aspects of image data can be discovered, and such aspects can better model the semantic of images. Specifically, we employ latent Dirichlet allocation (LDA) approach to cluster the image data into semantic topics and the distributions of image low-level features on such topics are used as the middle-level feature vectors of images. Meanwhile, a recently developed more efficient probabilistic representation of low-level features, i.e., Fisher Vector is used to complement the LDA representation for video concept detection. The experimental results on the TRECVID 2013 Semantic Indexing dataset have demonstrated the effectiveness of the proposed approach.