The main goal of the project is to define, build and test a prototype of a Context Aware TV System (CATVS). This system will rely on connected TV sets and/or STBs able to capture and elaborate the surrounding context in order to provide a new class of innovative and personalized services preserving the users’ privacy.
The context is defined as a set of entities and their status. One entity can be a person, a place or an object relevant for any type of interaction between user and device. Some parameters characterizing the context could be:
• location (e.g. where is the device: in the living room, in the kitchen, etc.);
• identity (e.g. who is the user, the user age, etc.);
• activity (e.g. what the user is doing: he/she is ironing, washing dishes, etc.);
• time (e.g. it is morning, it is evening, etc.);
• mood (e.g. the user is happy, sad, etc);
CATVS acquires information about the context (video, audio, text, etc..) by a suitable set of devices (cam, microphone, tablet, etc…).
On the basis of the captured context, CATVS will profile the relevant entities in order to properly implement new Internet based services, such as:
• Personalized Advertisement service: On the basis of marks in the TV transport stream, specifying the start and end of the video advertisement spot, CATVS will use entities’ profiles to select suitable videos advertisement from those made available from operators and/or advertisements companies.
• Personalized parental control service: CATVS will be able to “filter” unsuitable contents for children and activate proper countermeasures (e.g. force the TV receiver in a password-protected stand-by status).
• Smart audience evaluation service: the information on the captured context can be used to define new metrics in order to evaluate the viewing ratings (e.g. not only the number of viewers, but also their sex, age groups, etc).
Services could be studied and developed both in case of collaborative operators (i.e. broadcaster, or telco), and not collaborative ones.
CATVS implements a crowd-sourcing platform allowing the matches of user-profiles and TV contents. In fact, CATV services rely on a general taxonomy for TV contents that is dynamically tuned by collecting user feedback.
The Idealist2018 project is funded by the European Commission under the Leading Industrial Enabling Technologies ICT theme, of Horizon2020 – Research & Innovation Framework Programme, under grant agreement no. 645216. Any opinions expressed in these pages are those of the author/organisation and do not necessarily reflect the views of the European Commission.