A self-organizing multiple-view representation of 3D objects - PubMed (original) (raw)

A self-organizing multiple-view representation of 3D objects

S Edelman et al. Biol Cybern. 1991.

Abstract

We explore representation of 3D objects in which several distinct 2D views are stored for each object. We demonstrate the ability of a two-layer network of thresholded summation units to support such representations. Using unsupervised Hebbian relaxation, the network learned to recognize ten objects from different viewpoints. The training process led to the emergence of compact representations of the specific input views. When tested on novel views of the same objects, the network exhibited a substantial generalization capability. In simulated psychophysical experiments, the network's behavior was qualitatively similar to that of human subjects.

PubMed Disclaimer

Similar articles

• Psychophysical and physiological evidence for viewer-centered object representations in the primate.
  Logothetis NK, Pauls J. Logothetis NK, et al. Cereb Cortex. 1995 May-Jun;5(3):270-88. doi: 10.1093/cercor/5.3.270. Cereb Cortex. 1995. PMID: 7613082
• How are three-dimensional objects represented in the brain?
  Bülthoff HH, Edelman SY, Tarr MJ. Bülthoff HH, et al. Cereb Cortex. 1995 May-Jun;5(3):247-60. doi: 10.1093/cercor/5.3.247. Cereb Cortex. 1995. PMID: 7613080 Review.
• View-dependent object recognition by monkeys.
  Logothetis NK, Pauls J, Bülthoff HH, Poggio T. Logothetis NK, et al. Curr Biol. 1994 May 1;4(5):401-14. doi: 10.1016/s0960-9822(00)00089-0. Curr Biol. 1994. PMID: 7922354
• Forming sparse representations by local anti-Hebbian learning.
  Földiák P. Földiák P. Biol Cybern. 1990;64(2):165-70. doi: 10.1007/BF02331346. Biol Cybern. 1990. PMID: 2291903
• View-invariant object category learning, recognition, and search: how spatial and object attention are coordinated using surface-based attentional shrouds.
  Fazl A, Grossberg S, Mingolla E. Fazl A, et al. Cogn Psychol. 2009 Feb;58(1):1-48. doi: 10.1016/j.cogpsych.2008.05.001. Epub 2008 Jul 23. Cogn Psychol. 2009. PMID: 18653176 Review.

Cited by

• Using eye-tracking to parse object recognition: Priming activates primarily a parts-based but also a late-emerging features-based representation.
  Gerhardstein P, Olsen S. Gerhardstein P, et al. Atten Percept Psychophys. 2020 Aug;82(6):3096-3111. doi: 10.3758/s13414-020-02040-z. Atten Percept Psychophys. 2020. PMID: 32394068
• The haptic recognition of geometrical shapes in congenitally blind and blindfolded adolescents: is there a haptic prototype effect?
  Theurel A, Frileux S, Hatwell Y, Gentaz E. Theurel A, et al. PLoS One. 2012;7(6):e40251. doi: 10.1371/journal.pone.0040251. Epub 2012 Jun 28. PLoS One. 2012. PMID: 22761961 Free PMC article.
• Perception of fragmented images of three-dimensional objects as the observation angle changes.
  Chikhman VN, Shelepin YE, Foreman N, Passmore P. Chikhman VN, et al. Neurosci Behav Physiol. 2010 Jun;40(5):565-72. doi: 10.1007/s11055-010-9300-x. Epub 2010 May 14. Neurosci Behav Physiol. 2010. PMID: 20464503
• Effects of temporal association on recognition memory.
  Wallis G, Bülthoff HH. Wallis G, et al. Proc Natl Acad Sci U S A. 2001 Apr 10;98(8):4800-4. doi: 10.1073/pnas.071028598. Epub 2001 Apr 3. Proc Natl Acad Sci U S A. 2001. PMID: 11287633 Free PMC article.
• A template-matching pandemonium recognizes unconstrained handwritten characters with high accuracy.
  Larsen A, Bundesen C. Larsen A, et al. Mem Cognit. 1996 Mar;24(2):136-43. doi: 10.3758/bf03200876. Mem Cognit. 1996. PMID: 8881318

References

1. 1. Hum Neurobiol. 1985;4(4):219-27 - PubMed
2. 1. Nature. 1990 Jan 18;343(6255):263-6 - PubMed
3. 1. Nature. 1985 Sep 26-Oct 2;317(6035):314-9 - PubMed
4. 1. Science. 1990 Feb 23;247(4945):978-82 - PubMed
5. 1. Cognition. 1989 Aug;32(3):193-254 - PubMed

Publication types

MeSH terms