Reconstructing the World in Six Days
Authors:
- 1 June 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636919,p. 3287-3295
Abstract
We propose a novel, large-scale, structure-from-motion framework that advances the state of the art in data scalability from city-scale modeling (millions of images) to world-scale modeling (several tens of millions of images) using just a single computer. The main enabling technology is the use of a streaming-based framework for connected component discovery. Moreover, our system employs an adaptive, online, iconic image clustering approach based on an augmented bag-of-words representation, in order to balance the goals of registration, comprehensiveness, and data compactness. We demonstrate our proposal by operating on a recent publicly available 100 million image crowd-sourced photo collection containing images geographically distributed throughout the entire world. Results illustrate that our streaming-based approach does not compromise model completeness, but achieves unprecedented levels of efficiency and scalability.Keywords
This publication has 19 references indexed in Scilit:
- Street View Motion-from-Structure-from-Motion2013
- Towards Linear-Time Incremental Structure from Motion2013
- MatchMiner: Efficient Spanning Structure Mining in Large Image Collections2012
- Image webs: Computing and exploiting connectivity in image collections2010
- Building Rome in a day2009
- Modeling and Recognition of Landmark Image Collections Using Iconic Scene Graphs2008
- Towards Wiki-based Dense City Modeling2007
- Streaming Meshes2006
- Scalable Recognition with a Vocabulary Tree2006
- Recent developments on direct relative orientationISPRS Journal of Photogrammetry and Remote Sensing, 2006