Action Recognition with Actons
Authors:
- Jun ZhuJZJun Zhu
- Baoyuan WangBWBaoyuan Wang
- Xiaokang YangXYXiaokang Yang
- Wenjun ZhangWZWenjun Zhang
- Zhuowen TuZTZhuowen Tu
- 1 December 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15505499,p. 3559-3566
Abstract
With the improved accessibility to an exploding amount of video data and growing demands in a wide range of video analysis applications, video-based action recognition/classification becomes an increasingly important task in computer vision. In this paper, we propose a two-layer structure for action recognition to automatically exploit a mid-level ``acton'' representation. The actons are learned via a new max-margin multi-channel multiple instance learning framework. The learned actons (with no requirement for detailed manual annotations) thus observe a property of being compact, informative, discriminative, and easy to scale. This is different from the standard unsupervised (e.g. k-means) or supervised (e.g. random forests) coding strategies in action recognition. Applying the learned actons in our two-layer structure yields the state-of-the-art classification performance on Youtube and HMDB51 datasets.Keywords
This publication has 21 references indexed in Scilit:
- Harvesting Mid-level Visual Concepts from Large-Scale Internet Images2013
- Discovering discriminative action parts from mid-level video representations2012
- Image Classification by Hierarchical Spatial Pooling with Partial Least Squares Analysis2012
- Learning hierarchical invariant spatio-temporal features for action recognition with independent subspace analysis2011
- Recognizing human actions by attributes2011
- Maximum Margin Multiple Instance Clustering With Applications to Image and Text ClusteringIEEE Transactions on Neural Networks, 2011
- Actions in context2009
- Learning realistic human actions from movies2008
- Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories2006
- The Concave-Convex ProcedureNeural Computation, 2003