With the advent of the Internet of Things and the boost of Data Sciences, new challenges arise in human activity analysis, as well as a vast range of applications, especially in the domain of indoor environments. However, even if sensing devices (e.g. cameras, depth sensors, wearable devices, web access points) and computational power have improved significantly in the last years, the most promising algorithms for human activity analysis and event prediction are constrained by limited datasets, usage contexts, physical obstacles, personalized patterns of behaviors, etc. Another big challenge in data analysis and efficient application of machine learning arises from the fact that indoor environments impose problems related to sensor noise, scene clutter and false alarms attributed to contextual constraints. The scope of this workshop is to bring together researchers and developers working in the area of human activity analysis and event prediction in indoor environments and present state-of-the-art techniques and results in different domains where surveillance can bring major societal impact (health, security, resource management, building space management, etc.). The workshop will leverage results from the ICT4Life EU funded project and it will welcome research contributions from the broader research community.
Research papers are solicited in, but not limited to, the following areas topics:
- Human activity and behaviour analysis in indoor environments for security, health, space management, etc.
- Human activity monitoring in public spaces
- User re-identification and personalization
- Personality factors in human activity understanding
- Multi-modal fusion for human behaviour and activity analysis
- Anomaly detection in indoor activities