Structured Light Stereoscopic Imaging with （结构光立体成像）.pdfVIP

Structured Light Stereoscopic Imaging with Dynamic Pseudo-random Patterns Pierre Payeur and Danick Desjardins University of Ottawa, SITE, 800 King Edward, Ottawa, ON, Canada, K1N 6N5 {ppayeur,ddesjard}@site.uottawa.ca Abstract. Structured light stereoscopic imaging offers an efficient and afford- able solution to 3D modeling of objects. The majority of structured light pat- terns that have been proposed either provide a limited resolution or are sensitive to the inherent texture on the surface of the object. This paper proposes an in- novative imaging strategy that accomplishes 3D reconstruction of objects using a combination of spatial-neighboring and time-multiplexing structured light pat- terns encoded with uniquely defined pseudo-random color codes. The approach is extended with the concept of dynamic patterns that adaptively increases the reconstruction resolution. Original techniques are introduced to recover and validate pseudo-random codes from stereoscopic images, and to consistently map color and texture over the reconstructed surface map. Experimental results demonstrate the potential of the solution to create reconstructions with various densities of points and prove the robustness of the approach on objects with dif- ferent surface properties. Keywords: stereoscopic vision, structured light, 3D modeling, pseudo-random patterns. 1 Introduction The objective of the present work is to develop an integrated 3D imaging and recon- struction system that operates from affordable off-the-shelf equipment, provides the possibility to scan at various resolutions on demand, and operates on a wide variety of