Learning recognition and segmentation of 3-D objects from 2-D images

John J. Weng, N. Ahuja, T. S. Huang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A framework called Cresceptron is introduced for automatic algorithm design through learning of concepts and rules, thus deviating from the traditional mode in which humans specify the rules comprising a vision algorithm. With the Cresceptron, humans as designers need only to provide a good structure for learning, but they are relieved of most design details. The Cresceptron has been tested on the task of visual recognition: recognizing 3-D general objects from 2-D photographic images of natural scenes and segmenting the recognized objects from the cluttered image background. The Cresceptron uses a hierarchical structure to grow networks automatically, adaptively and incrementally through learning. The Cresceptron makes it possible to generalize training exemplars to other perceptually equivalent items. Experiments with a variety of real-world images are reported to demonstrate the feasibility of learning in the Cresceptron.

Original languageEnglish (US)
Title of host publication1993 IEEE 4th International Conference on Computer Vision
PublisherPubl by IEEE
Pages121-127
Number of pages7
ISBN (Print)0818638729
StatePublished - Jan 1 1993
Externally publishedYes
Event1993 IEEE 4th International Conference on Computer Vision - Berlin, Ger
Duration: May 11 1993May 14 1993

Publication series

Name1993 IEEE 4th International Conference on Computer Vision

Other

Other1993 IEEE 4th International Conference on Computer Vision
CityBerlin, Ger
Period5/11/935/14/93

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint

Dive into the research topics of 'Learning recognition and segmentation of 3-D objects from 2-D images'. Together they form a unique fingerprint.

Cite this