Scientific article
OA Policy
English

Probabilistic Motion Estimation Based on Temporal Coherence

Published inNeural computation, vol. 12, no. 8, p. 1839-1867
Publication date2000
Abstract

We develop a theory for the temporal integration of visual motion motivated by psychophysical experiments. The theory proposes that input data are temporally grouped and used to predict and estimate the motion flows in the image sequence. This temporal grouping can be considered a generalization of the data association techniques that engineers use to study motion sequences. Our temporal grouping theory is epxressed in terms of the Bayesian generalization of standard Kalman flitering. To implement the theory, we derive a parallel network that shares some properties of cortical networks. Computer simulations of this network demonstrate that our theory qualitatively accounts for psychophysical experiments on motion occlusion and motion outliers. In deriving our theory, we assumed spatial factorizability of the probability distributions and made the approximation of updating the marginal distributions of velocity at each point. This allowed us to perform local computations and simplified our implementation. We argue that these approximations are suitable for the stimuli we are considering (for which spatial coherence effects aer negligible).

Keywords
  • Kalman filter
  • Bayesian formulation
  • Motion flow field
  • Kolmogorov forward equation
  • Fokker-Plank equation
  • Temporal grouping
  • Temporal coherence
  • Psychophysical motion phenomena
  • Velocity estimation
  • Motion occlusion
  • Motion outlier
Affiliation entities Not a UNIGE publication
Citation (ISO format)
BURGI, Pierre-Yves, YUILLE, Alan L., GRZYWACZ, Norberto M. Probabilistic Motion Estimation Based on Temporal Coherence. In: Neural computation, 2000, vol. 12, n° 8, p. 1839–1867. doi: 10.1162/089976600300015169
Main files (1)
Article (Published version)
accessLevelPublic
Identifiers
Journal ISSN0899-7667
620views
346downloads

Technical informations

Creation13/11/2011 01:51:00
First validation13/11/2011 01:51:00
Update time14/03/2023 18:04:09
Status update14/03/2023 18:04:09
Last indexation29/10/2024 19:34:16
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack