Recognizing Human Actions: A Local SVM Approach
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 3 - Volume 03
A Bayesian Hierarchical Model for Learning Natural Scene Categories
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2
IEEE Transactions on Pattern Analysis and Machine Intelligence
Graph Embedding and Extensions: A General Framework for Dimensionality Reduction
IEEE Transactions on Pattern Analysis and Machine Intelligence
Behavior recognition via sparse spatio-temporal features
ICCCN '05 Proceedings of the 14th International Conference on Computer Communications and Networks
Unsupervised Learning of Human Action Categories Using Spatial-Temporal Words
International Journal of Computer Vision
Spatial-Temporal correlatons for unsupervised action classification
WMVC '08 Proceedings of the 2008 IEEE Workshop on Motion and video Computing
Image Categorization Based on a Hierarchical Spatial Markov Model
CAIP '09 Proceedings of the 13th International Conference on Computer Analysis of Images and Patterns
Unsupervised approximate-semantic vocabulary learning for human action and video classification
Pattern Recognition Letters
| Hi-index | 0.00 |
This paper presents a novel contextual spectral embedding (CSE) framework for human action recognition, which automatically learns the high-level features (motion semantic vocabulary) from a large vocabulary of abundant mid-level features (i.e. visual words). Our novelty is to exploit the inter-video context between mid-level features for spectral embedding, while the context is captured by the Pearson product moment correlation between mid-level features instead of Gaussian function computed over the vectors of point-wise information as mid-level feature representation. Our goal is to embed the mid-level features into a semantic low-dimensional space, and learn a much compact semantic vocabulary upon the CSE framework. Experiments on two action datasets demonstrate that our approach can achieve significantly improved results with respect to the state of the arts.