cs@cs.ucy.ac.cy | +357-22-892700

| | | | MyCS Portal |

SIMPOL VR - Synthesis of Dynamic Characters with motion capture data for human figure animation: Educating the police force

SIMPOL VR - Synthesis of Dynamic Characters with motion capture data for human figure animation: Educating the police force

Duration: 24 months (2009 - 2011)

Start: 1/12/2009

Principal Investigator: Γιώργος Χρυσάνθου

Main Funding Source: Cyprus Research Promotion Foundation

Total Cost: 37,440

Computer Animation, has advanced significantly as well as spectacularly over the last two-to-three decades. In particular, Computer Animation of human figure locomotion is becoming more and more popular in the recent years with virtual humans being used extensively in three-dimensional simulation environments. Such computer animation can be used in an extensive range of emerging applications, worldwide, for example in computer games, film industry, multimedia, training, education, medicine, virtual reality, architecture, security etc. The major aspiration of animating human figure characters is the continuous movement towards more responsive, controllable and thus believable characters for interactive virtual worlds.

A number of different techniques for animating (modelling plus motion of) the human body as well as the face have been developed, each with its own advantages but at the same time its own drawbacks. For animation of human figures, this work proposes a hybrid method which is based on the combination of motion capture-driven models and dynamic equations for motion control of articulated bodies. By using such a combination it is believed that the advantages of each method (naturalness of motion capture, control of dynamics) could be utilised while bypassing their drawbacks. However, the automatic transition from one method to the other, and perhaps back again, an unresolved problem, is a crucial part of the proposed work. These motions are going to be integrated with facial expressions of various emotions together with the realistic full-body modelling of the human avatars.

The results will be applied to the law enforcement community, for acquiring knowledge on principles and methods for curriculum and training, teaching and instruction for individuals and groups, as well as the measurement of their training effects according to the trainee’s responses. This will in essence be a pilot system, which will try to overcome the limitations of the training simulator the police force are currently using. Such a tool will help training officers when dealing with different scenarios that they experience in their everyday police-patrol excursions, always aiming at quick, easy and cost-effective solutions, where cost-effective would certainly mean safe responses without any spilling of blood.

For example, how to deal with situations when stopping a suspicious car that might end up being driven by a criminal, a drug addict or dealer, or illegally carrying a gun, or even driving under the influence of alcohol and how to react accordingly. This will be a 3D visualisation software development tool based on the actual movements (body language) and/or emotions (facial expressions) of the person under examination. A platform would be created for virtual reality settings, where different scenarios can be constructed by the trainer as well as changed dynamically according to the trainee's actions and responses.