Connectionists: Self-organization of spatio-temporal hierarchy in learning dynamic vision

[Jun Tani]

We announce a newly published paper which investigates how spatio-temporal
hierarchy can be self-organized in learning of dynamic visual patterns by
imposing multi-scales spatio-temporal constraints on neural activity in a
dynamic neural network model.

 

The following web page contains PDF, demonstrative video and source code.

http://neurorobot.kaist.ac.kr/project.html

 

Jung, M., Hwang, J., & Tani, J. (2015). Self-organization of spatio-temporal
hierarchy via learning of dynamic visual image patterns on action sequences.
PLoS One 10(7): e0131214. doi: 10.1371/journal.pone.0131214.

http://www.plosone.org/article/fetchObject.action?uri=info:doi/10.1371/journ
al.pone.0131214
<http://www.plosone.org/article/fetchObject.action?uri=info:doi/10.1371/jour
nal.pone.0131214&representation=PDF> &representation=PDF

 

Abstract:

It is well known that the visual cortex efficiently processes
high-dimensional spatial information by using a hierarchical structure.
Recently, computational models that were inspired by the spatial hierarchy
of the visual cortex have shown remarkable performance in image recognition.
Up to now, however, most biological and computational modeling studies have
mainly focused on the spatial domain and do not discuss temporal domain
processing of the visual cortex. Several studies on the visual cortex and
other brain areas associated with motor control support that the brain also
uses its hierarchical structure as a processing mechanism for temporal
information. Based on the success of previous computational models using
spatial hierarchy and temporal hierarchy observed in the brain, the current
report introduces a novel neural network model for the recognition of
dynamic visual image patterns based solely on the learning of exemplars.
This model is characterized by the application of both spatial and temporal
constraints on local neural activities, resulting in the self-organization
of a spatio-temporal hierarchy necessary for the recognition of complex
dynamic visual image patterns. The evaluation with the Weizmann dataset in
recognition of a set of prototypical human movement patterns showed that the
proposed model is significantly robust in recognizing dynamically occluded
visual patterns compared to other baseline models. Furthermore, an
evaluation test for the recognition of concatenated sequences of those
prototypical movement patterns indicated that the model is endowed with a
remarkable capability for the contextual recognition of long-range dynamic
visual image patterns.

 

 

Jun Tani, Ph.D

Professor, Department of Electrical Engineering, KAIST

Building: N1, room: 516

http://neurorobot.kaist.ac.kr/tani.htm

tani1216jp at gmail.com

042-350-7428

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20150715/47ba4209/attachment.html>