單一相機之同步定位與環境重建

線上場景重建的目標為同時追蹤相機所在的位置並且重建3D環境,而這樣的目的非常近似於影像同步定位與地圖建置( Visual SLAM )。目前有許多影像同步定位與地圖建置的方法可以提供非常高的準確度,不過他們多半需要立體視覺相機。若我們要使用單一相機來完成這個目標勢必會遇到更大的挑戰,但是單一相機具有較低的價格以及更容易被佈建等好處,使得單一相機有更高的吸引力。並且,環境3D重建相對於稀疏的3D地圖點在視覺呈現以及模型化上擁有更多的好處。因此,在本篇碩士論文中,我們著重在同步追蹤相機位置且進行3D環境重建。在本篇論文中,我們提出使用單一相機進行線上場景重建的演算法。我們透過最大事後機率(M.A.P.)與最佳化的估測來同時追蹤相機位置並建立密集的地圖點。我們也提出一個特徵點擴增的方法來增加地圖點的密度,接著使用延遲的方法來線上重建場景。並且, 我們使用所重建出的模型來進行不需要特徵擷取的影像定位(Visual Localization)。

關鍵字

線上場景重建；影像同步定位與地圖建置；單一相機；最佳化

並列摘要

The goal of online scene reconstruction is to simultaneously track the camera position and map the 3D environment. This is close to the spirit of vi- sual SLAM. There’re plenty of visual SLAM algorithms which can provide a high accuracy performance, but many of them rely on stereo cameras. It’s true that we’ll face much more challenges to accomplish this task with monocular camera. However, the advantages of cheaper price and easier adoption have made monocular approach more attractable. There are also many advantages on modeling and visualization of an environment with a physical scene re- construction instead of resorting to sparse 3D points. As a result, throughout this thesis, we focus on tracking the camera position and reconstruct the 3D scene simultaneously. In this thesis, we propose an online scene reconstruction algorithm with monocular camera. We apply a maximum a posteriori Bayesian approach with optimization technique to simultaneously track the camera and build a dense point cloud. We also propose a feature expansion method to expand the density of points, and then online reconstruct the scene with a delayed approach. Furthermore, we utilize the reconstructed model to accomplish visual localization task without extracting the features.

並列關鍵字

online scene reconstruction ； Visual SLAM ； monocular ； optimization

參考文獻

[2] J. Civera, A. J. Davison, and J. M. M. Montiel, ”Inverse Depth Parameterization for Monocular SLAM,” IEEE Transaction on Robotics, vol. 24, no. 5, 2008

[3] L. A. Clemente, A. J. Davison, I. D. Reid, J. Neira, and J. D. Tardo ́s, ”Mapping Large Loops with a Single Hand-Held Camera,” Robotics: Science and Systems, 2007

[4] P. Pinie ́s, and J. D. Tardo ́s, ”Large-Scale SLAM Building Conditionally Indepen- dent Local Maps: Application to Monocular Vision,” IEEE Transaction on Robotics, 2008

[6] M. Farrokhsiar, H. Najjaran, ”A Higher Order Rao-Blackwellized Particle Filter for Monocular vSLAM,” American Control Conference, 2010

[7] J. Kwon, and K. M. Lee, ”Monocular SLAM with Locally Planar Landmarks via Geometric Rao-Blackwellized Particle Filtering on Lie Groups,” IEEE Conference on Computer Vision and Pattern Recognition, 2010

國際替代計量

單一相機之同步定位與環境重建

全文下載

主題瀏覽