Dense and Scalable Reconstruction from Unstructured Videos with Occlusions

Wei, Jian; Resch, Benjamin; Lensch, Hendrik P. A.

Dense and Scalable Reconstruction from Unstructured Videos with Occlusions

dc.contributor.author	Wei, Jian	en_US
dc.contributor.author	Resch, Benjamin	en_US
dc.contributor.author	Lensch, Hendrik P. A.	en_US
dc.contributor.editor	Matthias Hullin and Reinhard Klein and Thomas Schultz and Angela Yao	en_US
dc.date.accessioned	2017-09-25T06:55:03Z
dc.date.available	2017-09-25T06:55:03Z
dc.date.issued	2017
dc.description.abstract	Depth-map-based multi-view stereo algorithms typically recover textureless surfaces by assuming smoothness per view, so they require processing different views to solve occlusions. Moreover, the highly redundant viewpoints of videos make exhaustive calculation of depth maps unfeasible for large scenes. This paper achieves dense and scalable reconstruction from videos by adaptively selecting a minimum subset of views from the unstructured camera paths, that are most beneficial for incremental occlusion handling and coverage improvement. Furthermore, we simplify and optimize each set of locally consistent points as the points accumulated from a cluster of previously processed views. By combining content-aware view selection and clustering, as well as cluster-wise point merging, our approach can reduce both computational and memory costs while producing accurate, concise, and dense 3D points, even for homogeneous areas. The superior efficiency and point-level fashion of our operations facilitate 3D modeling at large scales.	en_US
dc.description.sectionheaders	Shape Estimation and Analysis
dc.description.seriesinformation	Vision, Modeling & Visualization
dc.identifier.doi	10.2312/vmv.20171259
dc.identifier.isbn	978-3-03868-049-9
dc.identifier.pages	53-60
dc.identifier.uri	https://doi.org/10.2312/vmv.20171259
dc.identifier.uri	https://diglib.eg.org:443/handle/10.2312/vmv20171259
dc.publisher	The Eurographics Association	en_US
dc.subject	I.4.5 [Image Processing and Computer Vision]
dc.subject	Reconstruction
dc.subject
dc.subject	I.3.7 [Computer Graphics]
dc.subject	Three Dimensional Graphics and Realism
dc.subject	Color
dc.subject	Shading
dc.subject	Shadowing and Texture
dc.title	Dense and Scalable Reconstruction from Unstructured Videos with Occlusions	en_US