Unsupervised discovery of human activities from long‐time videos

Unsupervised discovery of human activities from long‐time videos

In this study, the authors propose a complete framework based on a hierarchical activity model to understand and recognise activities of daily living in unstructured scenes. At each particular time of a long‐time video, the framework extracts a set of space‐time trajectory features describing the gl...

Full description

Bibliographic Details
Main Authors: Salma Elloumi, Serhan Cosar, Guido Pusiol, Francois Bremond, Monique Thonnat
Format: Article
Language:English
Published: Wiley 2015-08-01
Series:IET Computer Vision
Subjects:
unsupervised discovery
hierarchical activity model
video recognition
space-time trajectory feature extraction
human motion information
perceptual feature chunk
Online Access:https://doi.org/10.1049/iet-cvi.2014.0311

Internet

https://doi.org/10.1049/iet-cvi.2014.0311

Similar Items