Multi-view 3D Reconstruction Techniques in Computer Vision
スポンサーリンク
概要
- 論文の詳細を見る
This paper gives introduction to 3D reconstruction techniques from multiple photographs in Computer Vision, a field known as 3D photography. The field of 3D photography has made dramatic progress in the last five years, and state-of-the-art techniques can now, for example, download millions of images of a city from Internet, find clusters of images that see the same parts of a city, estimate camera parameters, and recover their dense geometry. This paper gives high-level introduction to multi-view 3D reconstruction approaches and its three steps: 1) image acquisition; 2) camera parameter estimation; and 3) dense geometry reconstruction. The paper also introduces recent research projects and their reconstruction examples to illustrate the capabilities of state-of-the-art techniques.
- 2011-03-10