Abstract
In this paper we propose a novel method for recognizing human actions by exploiting a multi-layer representation based on a deep learning based architecture. A first level feature vector is extracted and then a high level representation is obtained by taking advantage of a Deep Belief Network trained using a Restricted Boltzmann Machine. The classification is finally performed by a feed-forward neural network. The main advantage behind the proposed approach lies in the fact that the high level representation is automatically built by the system exploiting the regularities in the dataset; given a suitably large dataset, it can be expected that such a representation can outperform a hand-design description scheme. The proposed approach has been tested on two standard datasets and the achieved results, compared with state of the art algorithms, confirm its effectiveness.
Original language | English |
---|---|
Title of host publication | 11th IEEE International Conference on Advanced Video and Signal-Based Surveillance, AVSS 2014 |
Place of Publication | Piscataway, NJ |
Publisher | IEEE |
Pages | 93-98 |
Number of pages | 6 |
ISBN (Electronic) | 978-1-4799-4871-0 |
ISBN (Print) | 978-1-4799-4870-3 |
DOIs | |
Publication status | Published - 8 Oct 2014 |
Externally published | Yes |
Event | 11th IEEE International Conference on Advanced Video and Signal-Based Surveillance, AVSS 2014 - Seoul, Korea, Republic of Duration: 26 Aug 2014 → 29 Aug 2014 Conference number: 11 |
Conference
Conference | 11th IEEE International Conference on Advanced Video and Signal-Based Surveillance, AVSS 2014 |
---|---|
Abbreviated title | AVSS |
Country/Territory | Korea, Republic of |
City | Seoul |
Period | 26/08/14 → 29/08/14 |