Most research efforts in scene analysis have concentrated on the analysis of block-world scenes. Having developed a good understanding of this limited world of computer vision, researchers are now trying to make computers see curved objects also. This paper presents an overview of the techniques developed for segmentation, representation, and recognition of curved objects in two-dimensional images and in three-dimensional scenes. The possible future directions of research are also discussed.
No takes yet. Share an insight, caveat, or question.
Jain et al. (1979) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: