摘要: Perceptual processes responsible for computing shape from several cues, including shading, texture, contour, and stereo, are examined. It is noted that these computational problems, as well of shaping motion, ill-posed in the sense Hadamard. suggested regularization theory can be used along with a priori knowledge to restrict space possible solutions, thus restore problem's well-prosedness. Some alternative methods outlined, idea active vision explored briefly connection problem. >
sci-hub.se 参考文章(63)
M. J. Brooks, Surface-normals from closed paths international joint conference on artificial intelligence. pp. 98- 101 ,(1979)
John R. Kender, Earl M. Smith, Shape from darkness: deriving surface information from dynamic shadows national conference on artificial intelligence. pp. 664- 667 ,(1986)
Michael Kass, Linear image features in stereopsis national conference on artificial intelligence. ,vol. 1, pp. 707- 713 ,(1986) , 10.1007/BF00133572
T. Poggio, MIT progress in understanding images Proceedings of a workshop on Image understanding workshop. pp. 56- 74 ,(1989)
John R. Kender, Shape from texture: an aggregation transform that maps a class of textures into surface orientation international joint conference on artificial intelligence. pp. 475- 480 ,(1979)
Berthold K. P. Horn, Michael J. Brooks, Shape and source from shading international joint conference on artificial intelligence. pp. 932- 936 ,(1985)
Franc Solina, Ruzena Bajcsy, Three dimensional object representation revisited The Computer Society of the IEEE. ,(1987)
H. Harlyn Baker, Thomas O. Binford, Depth from edge and intensity based stereo international joint conference on artificial intelligence. pp. 631- 636 ,(1981)
D. L. Brito, Michael D. Intriligator, Formal Models of Arms Races ,(1996)
Robert J. Woodham, Analysing images of curved surfaces Artificial Intelligence. ,vol. 17, pp. 117- 140 ,(1981) , 10.1016/0004-3702(81)90022-9