| Paper: | MA-P4.3 |
| Session: | Biometrics II: Human Activity, Gait, Gaze Analysis |
| Time: | Monday, September 17, 09:50 - 12:30 |
| Presentation: |
Poster
|
| Title: |
REAL-TIME AUTOMATIC DETECTION OF VIOLENT-ACTS BY LOW-LEVEL COLOUR VISUAL CUES |
| Authors: |
Alessandro Mecocci; University of Siena | | |
| | Francesco Micheli; University of Siena | | |
| Abstract: |
In this paper we address the specific problem of automatically detecting violent interpersonal acts in monocular colour video streams. Only little knowledge is assumed about the acquisition setup and light suppositions are made about the content of the acquired scenes. So the proposed approach is suitable in a wide range of practical cases. High reliability and general-purpose applicability is achieved by analysing low-level features (like the spatio-temporal behaviour of coloured stains), and by measuring some warping parameters. In this way it is not necessary to extract accurate target silhouettes, that is a critical task because of occlusions and overcrowding that are typical during interpersonal contacts. A suitable index called Maximum Warping Energy (MWE) has been defined to describe the localized spatial-temporal complexity of colour conformations. |
