期刊
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION
卷 32, 期 -, 页码 63-73出版社
ACADEMIC PRESS INC ELSEVIER SCIENCE
DOI: 10.1016/j.jvcir.2015.07.016
关键词
Human interaction recognition; Activity recognition; Multiple instance learning; Video retrieval; Video analysis; Human actions; Human interactions; Spatial embedding
资金
- Scientific and Technological Research Council of Turkey (TUBITAK) [112E149]
In this work, we look into the problem of recognizing two-person interactions in videos. Our method integrates multiple visual features in a weakly supervised manner by utilizing an embedding-based multiple instance learning framework. In our proposed method, first, several visual features that capture the shape and motion of the interacting people are extracted from each detected person region in a video. Then, two-person visual descriptors are formed. Since the relative spatial locations of interacting people are likely to complement the visual descriptors, we propose to use spatial multiple instance embedding, which implicitly incorporates the distances between people into the multiple instance learning process. Experimental results on two benchmark datasets validate that using two-person visual descriptors together with spatial multiple instance learning offers an effective way for inferring the type of the interaction. (C) 2015 Elsevier Inc. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据