Practical time bundle adjustment for 3d reconstruction on the GPU

Authors:
Siddharth Choudhary;Shubham Gupta;P. J. Narayanan
Affiliations:
Center for Visual Information Technology, International Institute of Information Technology, Hyderabad, India;Center for Visual Information Technology, International Institute of Information Technology, Hyderabad, India;Center for Visual Information Technology, International Institute of Information Technology, Hyderabad, India
Venue:
ECCV'10 Proceedings of the 11th European conference on Trends and Topics in Computer Vision - Volume Part II
Year:
2010

Citing 8
Cited 0

Bundle Adjustment - A Modern Synthesis

ICCV '99 Proceedings of the International Workshop on Vision Algorithms: Theory and Practice
Unsupervised 3D Object Recognition and Reconstruction in Unordered Datasets

3DIM '05 Proceedings of the Fifth International Conference on 3-D Digital Imaging and Modeling
OpenVIDIA: parallel GPU computer vision

Proceedings of the 13th annual ACM international conference on Multimedia
Photo tourism: exploring photo collections in 3D

ACM SIGGRAPH 2006 Papers
NVIDIA Tesla: A Unified Graphics and Computing Architecture

IEEE Micro
SBA: A software package for generic sparse bundle adjustment

ACM Transactions on Mathematical Software (TOMS)
A parallel levenberg-marquardt algorithm

Proceedings of the 23rd international conference on Supercomputing
Building Rome on a cloudless day

ECCV'10 Proceedings of the 11th European conference on Computer vision: Part IV

Quantified Score

Hi-index	0.00

Visualization

Abstract

Large-scale 3D reconstruction has received a lot of attention recently. Bundle adjustment is a key component of the reconstruction pipeline and often its slowest and most computational resource intensive. It hasn't been parallelized effectively so far. In this paper, we present a hybrid implementation of sparse bundle adjustment on the GPU using CUDA, with the CPU working in parallel. The algorithm is decomposed into smaller steps, each of which is scheduled on the GPU or the CPU. We develop efficient kernels for the steps and make use of existing libraries for several steps. Our implementation outperforms the CPU implementation significantly, achieving a speedup of 30-40 times over the standard CPU implementation for datasets with upto 500 images on an Nvidia Tesla C2050 GPU.