Inferring Motion of Cylindrical Object From Shape Information

Abstract

This paper explores methods for Inferring motion · of a cylindrical · object from an Image sequence. A local shading analysis method segments • each Image Into spherical, cylindrical, and planar 'surfaces. The cylindrical surface Is characterized with a dlr:ectlon of l generating lines determined from I spatial derivatives In the Image. An extended reflectance map method • Is applicab'e to 'estimate gradients of base In consecutive frames If the .' lighting conditions are known. Another approach to lnf erring of motion Is to use shape Information which needs an additional ! constraint. If two planes of the object are not parallel, then , the finding' ofcorrespondence between frames Is easy. For the case ofparallel . planes, we Introduce · an addltior\al assumption thatthe planes are orthogonal to the generating lines. This strong : constraint ' Is not sufficient for determhilng the surface normals from single view. The . orthogonality also gives another constraint between a change In area of the base plane and that In \ length of generating line from the first to the second · frames. Thus Iwe can estimate orientations , of the base plane from two views. i Finally, motion parameters are estimated by correlating two shapes of the base boundary mapped :from the two views.