Clustering learning for robotic vision

Abstract

We present the clustering learning technique applied to multi-layer feedforward
\ndeep neural networks. We show that this unsupervised learning technique can
\ncompute network filters with only a few minutes and a much reduced set of parameters.
\nThe goal of this paper is to promote the technique for general-purpose
\nrobotic vision systems. We report its use in static image datasets and object tracking
\ndatasets. We show that networks trained with clustering learning can outperform
\nlarge networks trained for many hours on complex datasets.