Learning joint top-down and bottom-up processes for 3D visual inference

Cristian Sminchisescu, Atul Kanaujia, Dimitris Metaxas

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

62 Scopus citations

Abstract

We present an algorithm for jointly learning a consistent bidirectional generative-recognition model that combines top-down and bottom-up processing for monocular 3d human motion reconstruction. Learning progresses in alternative stages of self-training that optimize the probability of the image evidence: the recognition model is tunned using samples from the generative model and the generative model is optimized to produce inferences close to the ones predicted by the current recognition model. At equilibrium, the two models are consistent. During on-line inference, we scan the image at multiple locations and predict 3d human poses using the recognition model. But this implicitly includes one-shot generative consistency feedback. The framework provides a uniform treatment of human detection, 3d initialization and 3d recovery from transient failure. Our experimental results show that this procedure is promising for the automatic reconstruction of human motion in more natural scene settings with background clutter and occlusion.

Original languageEnglish (US)
Title of host publicationProceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006
Pages1743-1750
Number of pages8
DOIs
StatePublished - 2006
Event2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006 - New York, NY, United States
Duration: Jun 17 2006Jun 22 2006

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume2
ISSN (Print)1063-6919

Other

Other2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006
Country/TerritoryUnited States
CityNew York, NY
Period6/17/066/22/06

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Learning joint top-down and bottom-up processes for 3D visual inference'. Together they form a unique fingerprint.

Cite this