SURveillance Video Archives iNvestigation assisTant

SURVANT will deliver an innovative system that will collect the relevant videos from heterogeneous repositories, extract inter/intra-camera video analytics, enrich the analytics using reasoning and inference technologies, and offer a unified search interface to the user. An intuitive interface with a relaxed learning curve will assist the user create accurate search queries and receive the results using advanced visualization tools. Ethical management of personal data collected from surveillance videos is integrated in the system design.

SURVANT aims to develop a product that will introduce solutions beyond the industry state-of-the-art to face the challenges identified in the markets targeted, namely:

• Scalability: SURVANT will address system scalability issues that emerge from the explosion in the amount of available video content.
• Video analysis: SURVANT will perform video analysis employing Deep Learning (DL) techniques.
• Event Enrichment and Reasoning: SURVANT will deliver an inference framework able to combine low-level information and semantic annotations to enable automated reasoning mechanisms.
• Video and Image indexing: SURVANT will deal with large scale image/video retrieval application by efficiently indexing the content extracted from large image/video repositories.
• Intelligent GIS-based user interface: A GIS-based user interface will be developed and demonstrated, allowing the user to execute targeted queries, in terms of location and time, and visualize the returned results in analytical or summarized forms.
• Legal and ethical issues: SURVANT will deliver an intelligent and extensible framework for automated video analysis and analytics respectful of human rights according to European legal framework for privacy and data protection.

Key functionalities have been identified that will provide high added value for potential customers compared to the competition are:

• Situational awareness framework,
• Advanced content-based search, and
• Search expansion tools.

VCL is responsible for the technical management of the project. It develops Deep Learning methodologies for image and video analysis, to detect objects, actions and events in the examined content. VCL also leads complex query formulation and has an active role in event enrichment, reasoning, data protection and privacy assurance.