Statistical correlations between two-dimensional images and three-dimensional structures in natural scenes

Abstract

In spite of the recent surge in the popularity of statistical approaches to vision, the joint statistics of coregistered range and light-intensity images have gone relatively unexplored. We investigate statistical correlations between images and the surface shapes that produced them. We determine which linear properties of range images can be best predicted from simple computations on intensity information, and we determine those properties of intensity images that best predict range information. We find that significant (up to ϭ 0.45) and potentially exploitable correlations exist between linear properties of range and intensity images, and we explore the structure of these correlations.

Statistical Analysis of Second-Order Relations of 3d Structures

“…Moreover, with the advances in technology, it has been also possible to analyze the underlying 3D world using 3D range scanners (Howe and Purves, 2004;Huang et al, 2000;Potetz and Lee, 2003;Yang and Purves, 2003).…”

“…There have been only a few studies that have investigated the 3D world from range data (Howe and Purves, 2004;Huang et al, 2000;Kalkan et al, 2006;Potetz and Lee, 2003;Yang and Purves, 2003). In (Yang and Purves, 2003), the distribution of roughness, size, distance, 3D orientation, curvature and independent components of surfaces was analyzed.…”

Statistical Analysis of Local 3D Structure in 2D Images

“…Moreover, with the advances in technology, it has been also possible to analyze the underlying 3D world using 3D range scanners [10,11,19,27].…”

“…There have been only a few studies that have analyzed the 3D world from range data [10,11,19,27]. In [27], the distribution of roughness, size, distance, 3D orientation, curvature and independent components of surfaces was analyzed.…”

“…They showed that range images are much simpler to analyze than optical images and that a 3D scene is composed of piecewise smooth regions. In [19], the correlation between light intensities of the image data and the corresponding range data as well as surface convexity were investigated. They could justify the event that brighter objects are closer to the viewer, which is used by shape from shading algorithms in estimating depth.…”