Modeling people and places with internet photo collections
Communications of the ACM
Modeling People and Places with Internet Photo Collections
Queue - Networks
Worldwide pose estimation using 3d point clouds
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part I
Improving image-based localization by active correspondence search
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part I
Covariance propagation and next best view planning for 3d reconstruction
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part II
Visibility probability structure from sfm datasets and applications
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part V
Towards fast image-based localization on a city-scale
Proceedings of the 15th international conference on Theoretical Foundations of Computer Vision: outdoor and large-scale real-world scene analysis
Video enhancement leveraging high-quality depth maps
Proceedings of the 11th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry
A Self-adaptive ASIFT-SH method
Advanced Engineering Informatics
Simultaneous multiple rotation averaging using lagrangian duality
ACCV'12 Proceedings of the 11th Asian conference on Computer Vision - Volume Part III
Large-Scale bundle adjustment by parameter vector partition
ACCV'12 Proceedings of the 11th Asian conference on Computer Vision - Volume Part IV
Exploring high-level plane primitives for indoor 3d reconstruction with a hand-held RGB-D camera
ACCV'12 Proceedings of the 11th international conference on Computer Vision - Volume 2
Fine-grained semi-supervised labeling of large shape collections
ACM Transactions on Graphics (TOG)
FOCUS: clustering crowdsourced videos by line-of-sight
Proceedings of the 11th ACM Conference on Embedded Networked Sensor Systems
Fast iterative graph computation with block updates
Proceedings of the VLDB Endowment
| Hi-index | 0.02 |
Recent work in structure from motion (SfM) has successfully built 3D models from large unstructured collections of images downloaded from the Internet. Most approaches use incremental algorithms that solve progressively larger bundle adjustment problems. These incremental techniques scale poorly as the number of images grows, and can drift or fall into bad local minima. We present an alternative formulation for SfM based on finding a coarse initial solution using a hybrid discrete-continuous optimization, and then improving that solution using bundle adjustment. The initial optimization step uses a discrete Markov random field (MRF) formulation, coupled with a continuous Levenberg-Marquardt refinement. The formulation naturally incorporates various sources of information about both the cameras and the points, including noisy geotags and vanishing point estimates. We test our method on several large-scale photo collections, including one with measured camera positions, and show that it can produce models that are similar to or better than those produced with incremental bundle adjustment, but more robustly and in a fraction of the time.