People over 50 will rise by 35% between 2005 and 2050. The number of people over 85 will triple by 2050. OECD analyses forecast increasing costs as a result of ageing populations. Thus, attention to the needs of the elderly and disabled is today in all developed countries one of the great challenges of social and economic policies. The challenge is to serve people who, by being in vulnerable situations, need support to develop the most essential activities of daily living.Under these requirements, it raises the question how ICTs could be applied to the environments in which the users are, allowing to interact with people in a natural way, wherever they are needed, sensitive to the user and his context and act proactively.So, there is a worldwide interest in research and development of systems for the analysis of people’s activities, especially those most in need, elderly and disabled. These systems are composed of networks of sensors that can analyse the environment to extract knowledge from it in order to detect anomalous behaviour or send alarms to care services.Vision systems for behaviour analysis have spread in recent years, mainly by security demands and the reducing cost of the devices. However, most systems are applied to outdoor environments and, when they are installed indoor they are used in large facilities, in few cases within the home. This is due to two facts: necessity of multi-camera distributed systems, and privacy requirements from users.This project will deal with these aspects, designing and developing intelligent multimodal systems for the behaviour analysis of people in private environments, especially in their homes. People would accept these technology and services if we can ensure their privacy under any circumstance. So, techniques will be developed to ensure the privacy of those being studied. Information richness of these visual devices would open a new field of services for support people, but also their families, carers...