The Recognition of Human Movement Using Temporal Templates
IEEE Transactions on Pattern Analysis and Machine Intelligence
Recognizing Action at a Distance
ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Recognizing Human Actions: A Local SVM Approach
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 3 - Volume 03
ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2
Behavior recognition via sparse spatio-temporal features
ICCCN '05 Proceedings of the 14th International Conference on Computer Communications and Networks
Fast Human Detection by Boosting Histograms of Oriented Gradients
ICIG '07 Proceedings of the Fourth International Conference on Image and Graphics
Unsupervised Learning of Human Action Categories Using Spatial-Temporal Words
International Journal of Computer Vision
Human Action Recognition by Semilatent Topic Models
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
In this study, we present a simple and effective approach to human action recognition in real time video. We use oriented gradients to represent human contour and recognize human action. First, we detect the human area and obtain human contour information. And then, we use the binary image of contour to extract features and divide the contour area of each frame into several blocks. In each block, we calculate the histogram of oriented gradients and extract the main orientation of gradients as the block feature. In each frame, features of all blocks are concatenated to one feature vector to represent human pose. Then, we concatenate human pose features of sequential frames as an action feature using a sliding window with overlapping ratio. After that, we train a SVM classifier by these action features. In the experiments, our approach has a good recognition performance compared with state-of-the-art methods.