This research introduces a conceptual framework for enforcing privacy-related restrictions in smart video surveillance systems based on danger levels and incident types to be handled. It increases the selectivity of surveillance by restricting data processing to individuals associated to incidents under investigation. Constraints are enforced by usage control, which is instantiated for video surveillance for the first time and enables tailoring such systems to comply with data protection law.

This research introduces a conceptual framework for enforcing privacy-related restrictions in smart video surveillance systems based on danger levels and incident types to be handled. It increases the selectivity of surveillance by restricting data processing to individuals associated to incidents under investigation. Constraints are enforced by usage control, which is instantiated for video surveillance for the first time and enables tailoring such systems to comply with data protection law. This work was published by Saint Philip Street Press pursuant to a Creative Commons license permitting commercial use. All rights not granted by the work's license are retained by the author or authors.

This work proposes a feature-based probabilistic data association and tracking approach (FBPDATA) for multi-object tracking. FBPDATA is based on re-identification and tracking of individual video image points (feature points) and aims at solving the problems of partial, split (fragmented), bloated or missed detections, which are due to sensory or algorithmic restrictions, limited field of view of the sensors, as well as occlusion situations.

The availability of video data is an opportunity and a challenge for law enforcement agencies. Face recognition methods can play a key role in the automated search for persons in the data. This work targets efficient representations of low-quality face sequences to enable fast and accurate face search. Novel concepts for multi-scale analysis, dataset augmentation, CNN loss function, and sequence description lead to improvements over state-of-the-art methods on surveillance video footage.

The functioning of intelligent autonomous systems requires constant situation awareness and cognition analysis. Thus, it needs a memory structure that contains a description of the surrounding environment (world model) and serves as a central information hub. This book presents a row of theoretical and experimental results in the field of world modeling. This includes areas of dynamic and prior knowledge modeling, information fusion, management and qualitative/quantitative information analysis.

This book proposes a novel deep learning based detection method, focusing on vehicle detection in aerial imagery recorded in top view. The base detection framework is extended by two novel components to improve the detection accuracy by enhancing the contextual and semantical content of the employed feature representation. To reduce the inference time, a lightweight CNN architecture is proposed as base architecture and a novel module that restricts the search area is introduced.

"This work proposes a Multibody Structure from Motion (MSfM) algorithm for moving object reconstruction that incorporates instance-aware semantic segmentation and multiple view geometry methods. The MSfM pipeline tracks two-dimensional object shapes on pixel level to determine object specific feature correspondences, in order to reconstruct 3D object shapes as well as 3D object motion trajectories" -- Publicaciones de Arquitectura y Arte.