CSISmartScan 3D




CSISmartScan3D will explore the possibilities and limitations for using modern sensor technologies with regard to requirements in the context of crime scene documentation, notably whether and to which extent adequate and complete 3D models can be captured. In the range of core methods, the potential of inter-modal sensor fusion will be of particular interest.

3D crime site scanning and documentation based on cost-efficient 3D sensors, tablets and smartphones. 3D models support the permanent and complete preservation of the spatial configuration at crime and accident scenes. They form a valuable basis for computer-aided case analysis, documentation and comprehensible visual case preparation and presentation in the context of court proceedings, leading to increased transparency, reproducibility and legal certainty in the end. Current 3D scanning technologies are unable to cope with all the requirements for general crime scene documentation. Factors such as model quality, acquisition time, disposability, disruption of crime scene investigations, ergonomics, and prohibitively high costs limit the use of 3D scanners to specialized questions apart from routine procedures used for overall crime site documentation. Furthermore, current 3D scanners do not support tasks beyond pure 3D model acquisition without direct benefit for crime site investigation. The resulting 3D models are only used for case analysis after post-processing. CSISmartScan3D will investigate the feasibility of a new kind of 3D crime scene documentation system beyond pure 3D model acquisition taking into account the requirements at the site. The models will be directly used as 3D references for other crime site investigation tasks. Sensor data fusion of multiple cost effective sensors and model visualization on the scanning device will lead to a high degree of robustness of the acquisition process, model quality, and completeness, whilst avoiding delays of other crime site investigations. The idea to use the model as a 3D reference model as well as the possibility to annotate, e.g. traces, will further increase efficiency. CSISmartScan3D will explore the possibilities and limitations of the abovementioned sensor technologies, mobile devices, and interaction techniques. Tests will show whether and to which extent these techniques are suited for the acquisition of adequate, complete annotated 3D models of the crime site.

Related Publications

Friedrich Fraundorfer and Fabian Schenk RESLAM: A real-time robust edge-based SLAM systemRESLAM: A real-time robust edge-based SLAM system Show publication in PURE
Friedrich Fraundorfer, Horst Bischof and Fabian Schenk UAV-Anwendungen im Bergbau - Modellversuche als Beitrag zur photogrammetrischen Forschung und EntwicklungUAV-Anwendungen im Bergbau - Modellversuche als Beitrag zur photogrammetrischen Forschung und Entwicklung30-43 Show publication in PURE
Friedrich Fraundorfer, Fabian Schenk and Rafael Jakob Weilharter Globally Consistent Dense Real-Time 3D Reconstruction from RGBD DataGlobally Consistent Dense Real-Time 3D Reconstruction from RGBD Data121-127 Show publication in PURE
Friedrich Fraundorfer and Fabian Schenk Combining Edge Images and Depth Maps for Robust Visual OdometryCombining Edge Images and Depth Maps for Robust Visual Odometry , Link Show publication in PURE
Friedrich Fraundorfer and Fabian Schenk Robust Edge-based Visual Odometry using Machine-Learned EdgesRobust Edge-based Visual Odometry using Machine-Learned Edges1297-1304 , Link Show publication in PURE
Friedrich Fraundorfer and Fabian Schenk CSISmartScan3D - Cost-efficient Integrated 3D Crime Scene Documentation Show publication in PURE
Friedrich Fraundorfer, Horst Bischof, Ludwig Mohr and Fabian Schenk Guided sparse camera pose estimationGuided sparse camera pose estimation139-146 , Link Show publication in PURE
Horst Bischof and Fabian Schenk Automated Segmentation of the Walkable Area from Aerial Images for Evacuation SimulationAutomated Segmentation of the Walkable Area from Aerial Images for Evacuation Simulation125-135 Show publication in PURE
Fabian Schenk Automatic high-speed video glottis segmentation using salient regions and 3D geodesic active contours Annals of the BMVA 2015, 1-15, 2015 , Link Show publication in PURE
Horst Bischof and Fabian Schenk Automatic glottis segmentation from laryngeal high-speed videos using 3D active contoursAutomatic glottis segmentation from laryngeal high-speed videos using 3D active contours111-116 Show publication in PURE
Friedrich Fraundorfer and Fabian Schenk Automatic Muck Pile Characterization from UAV ImagesAutomatic Muck Pile Characterization from UAV Images Show publication in PURE
Project Team
image/svg+xml