Most structure from motion work assume that tracking is a separate issue from recovery of structure and motion. This paper argues that in certain cases, it helps to use the original images to refine the tracks in the course of estimating structure and motion. We illustrate this by presenting two algorithms: one for appearance-based structure from complete rotation, and another for 3-D face mode...

We provide exact shape reconstruction formulas in the spirit of the Linear Sampling method for a class of inverse problems in shape determination in the context of timeindependent partial differential equations. To this end, we prove a general theorem how, and under which assumptions, domain characterizations based on the range of the square root of an operator transform into domain characteriz...

We present a model-based method for detailed automatic building reconstruction from images. The method starts with recovering a coarse building model consisting of the main object planes. This model is upgraded by detecting deviations from the planes and fitting predefined shape primitives to these deviations to refine the geometry. The proposed method is presented on real datasets.

To a layperson, a raindrop would have the shape of a tear drop, as often renditioned in popular media. However, to an acute individual, and as pointed out fervently by Fraser [17], a raindrop would have more of a spherical shape. The tear drop shape applies only just as the drop falls from a surface. It would then take a somewhat spherical form, and then distort to a shape like an oblate sphero...

Subjects imagined a designated two-dimensional shape rotating within a blank, circular field at a self-determined rate. At some point during the mental rotation, a test shape was presented at one of 12 picture-plane orientations, and the subject was required to determine as rapidly as possible whether the test shape was the same as the originally designated shape or was its mirror image. When t...

This paper describes a method to accelerate the generation of shape primitives for N-dimensional images N X . These shape primitives can be used in conditions for topology preserving erosion or skeletonization in Ndimensional images. The method is based on the possibility to describe primitives for intrinsic dimensions 1 N N = − % by quadratic equations of the form ( ) 1 2

We present a novel method to characterize 3D surfaces through the computation of a function called (multiscale) area projection transform, measuring the likelihood of points in the 3D space to be center of radial symmetry at selected scales (radii). The function is derived through a simple geometric framework based on parallel surfaces and can be easily computed on triangulated meshes. It measu...

Shape is an important aspects in recognizing plants. Several approaches have been introduced to identify objects, including plants. Combination of geometric features such as aspect ratio, compactness, and dispersion, or moments such as moment invariants were usually used toidentify plants. In this research, a comparative experiment of 4 methods to identify plants using shape features was accomp...