论文信息 - An Analog VLSI Chip for Thin-Plate Surface Interpolation

An Analog VLSI Chip for Thin-Plate Surface Interpolation

Reconstructing a surface from sparse sensory data is a well-known problem in computer vision. This paper describes an experimental analog VLSI chip for smooth surface interpolation from sparse depth data. An eight-node ID network was designed in 3µm CMOS and successfully tested. The network minimizes a second-order or "thin-plate" energy of the surface. The circuit directly implements the coupled depth/slope model of surface reconstruction (Harris, 1987). In addition, this chip can provide Gaussian-like smoothing of images.

John G. Harris | J. Harris

[1] W. G. Proctor,et al. Analogues for the Solution of Boundary Value Problems , 1967 .

[2] V. E. Bukhman,et al. Analogues for the solution of boundary-value problems , 1967 .

[3] W. Eric L. Grimson,et al. From images to surfaces , 1981 .

[4] Demetri Terzopoulos,et al. Multilevel computational processes for visual surface reconstruction , 1983, Comput. Vis. Graph. Image Process..

[5] T. Poggio,et al. Ill-posed problems in early vision: from computational theory to analogue networks , 1985, Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences.

[6] Alan L. Yuille,et al. A regularized solution to edge detection , 1985, J. Complex..

[7] Carver Mead,et al. Analog VLSI and neural systems , 1989 .