Learning stereo disparity using temporal smoothness constraints: A computational model

Author: Stone, James V.¹

Source: Spatial Vision, Volume 10, Number 1, 1996 , pp. 15-29(15)

Publisher: VSP, an imprint of Brill

Key:

- Free Content

- New Content

- Subscribed Content

- Free Trial Content

Abstract:

An unsupervised learning algorithm is presented for learning stereo disparity. A key assumption is that surface depth varies smoothly over time. This assumption is consistent with a learning rule which maximizes the long-term variance of each unit's outputs, whilst simultaneously minimizing its short-term variance. The learning rule involves a linear combination of anti-Hebbian and Hebbian weight changes, over short and long time scales, respectively. The model is demonstrated on a hyperacuity task: estimating sub-pixel stereo disparity from a temporal sequence of stereograms. The algorithm generalizes, without additional learning, to previously unseen image sequences.

Document Type: Research article

DOI: 10.1163/156856896X00033

Affiliations: 1: Department of Psychology, University of Sheffield, Western Bank, Sheffield, SIO 2UR, England

The full text electronic article is available for purchase. You will be able to download the full text electronic article after payment.

$25.00 plus tax Refund Policy