Real-time hybrid ToF multi-camera rig fusion system for depth map enhancement
2011 (English)In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, IEEE , 2011Conference paper (Refereed)
We present a full real-time implementation of a multilateral filtering system for depth sensor data fusion with 2-D data. For such a system to perform in real-time, it is necessary to have a real-time implementation of the filter, but also a real-time alignment of the data to be fused. To achieve an automatic data mapping, we express disparity as a function of the distance between the scene and the cameras, and simplify the matching procedure to a simple indexation procedure. Our experiments show that this implementation ensures the fusion of 3-D data and 2-D data in real-time and with high accuracy.
Place, publisher, year, edition, pages
IEEE , 2011.
automatic data mapping, depth map enhancement, multilateral filtering system, real-time hybrid ToF multicamera rig fusion system, sensor data fusion, Calibration, Cameras, Image edge detection, Image resolution, Integrated circuits, Real time systems, Table lookup
IdentifiersURN: urn:nbn:se:kth:diva-72532DOI: 10.1109/CVPRW.2011.5981740ScopusID: 2-s2.0-80054951743OAI: oai:DiVA.org:kth-72532DiVA: diva2:487731
2011 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2011, Colorado Springs, CO, 20-25 June, 2011
QC 201202082012-01-312012-01-312016-04-28Bibliographically approved