eprintid: 41384
rev_number: 15
eprint_status: archive
userid: 1903
dir: disk0/00/04/13/84
datestamp: 2016-07-10 07:29:59
lastmod: 2023-01-30 11:51:24
status_changed: 2023-01-30 11:51:24
type: conference_item
metadata_visibility: show
creators_name: Vidal, Esteban
creators_name: Piotto, Nicola
creators_name: Cordara, Giovanni
creators_name: Morán Burgos, Francisco
creators_id: fmb@gti.ssr.upm.es
title: Automatic video to point cloud registration in a structure-from- motion framework
ispublished: pub
subjects: telecomunicaciones
keywords: 3D Reconstruction, SfM, Video Registration, Point Cloud Alignment
abstract: In Structure-from-Motion (SfM) applications, the capability of integrating new visual information into existing 3D models is an important need. In particular, video streams could bring significant advantages, since they provide dense and redundant information, even if normally only relative to a limited portion of the scene. In this work we propose a fast technique to reliably integrate local but dense information from videos into existing global but sparse 3D models. We show how to extract from the video data local 3D information that can be easily processed allowing incremental growing, refinement, and update of the existing 3D models. The proposed technique has been tested against two state-of-the-art SfM algorithms, showing significant improvements in terms of computational time and final point cloud density.
date: 2015
date_type: published
publisher: IEEE
official_url: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7351282
full_text_status: public
pres_type: paper
publication: 2015 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)
pagerange: 2646-2650
event_title: International Conference on Image Processing (ICIP 2015)
event_location: Quebec City, Canada
event_dates: 27/09/2015 - 30/09/2015
event_type: conference
institution: Telecomunicacion
department: Senales
refereed: TRUE
issn: 1522-4880
book_title: International Conference on Image Processing (ICIP 2015)
rights: by-nc-nd
comprojects_type: FP7
comprojects_code: 610691
comprojects_acronym: BRIDGET
comprojects_leader: The University of Surrey
comprojects_title: BRIDging the Gap for Enhanced broadcasT
citation:   Vidal, Esteban and Piotto, Nicola and Cordara, Giovanni and Morán Burgos, Francisco  (2015).  Automatic video to point cloud registration in a structure-from- motion framework.  In: "International Conference on Image Processing (ICIP 2015)", 27/09/2015 - 30/09/2015, Quebec City, Canada. pp. 2646-2650.    
document_url: https://oa.upm.es/41384/1/INVE_MEM_2015_229726.pdf