[visionlist] Call for Papers Special Issue of Pattern Recognition Journal

[Terrance Boult]

*Call for Papers*

Pattern Recognition Special Issue on Open World Robust Pattern Recognition

*Aims and Scope*

Most traditional pattern recognition methods are based on the
closed-world assumption. However, in real-world applications, the
environment is usually open and dynamic, requiring new models and
algorithms to deal with the openness. When recontextualized into open
world recognition, many once solved pattern recognition tasks become
significant challenging tasks again. In open world, a pattern
recognition model should have the ability to reject out-of-distribution
and unknown samples. It is also necessary and important for the system
to be able to discover new classes, and then improve its knowledge with
continual learning, class-incremental learning, or lifelong learning.
How to avoid catastrophic forgetting in this process is a fundamental
problem. Moreover, in open world, it is shown that state-of-the-art
pattern recognition models (like deep neural networks) are easily fooled
by assigning high confidence predictions for unrecognizable or forged
images, indicating that although the accuracy is high, it is not as
robust as human vision when dealing with outliers and adversarial
attacks. The adversarial examples which add a small perturbation
(particularly designed) on the input sample may lead to unexpected or
incorrect predictions for pattern recognition systems, leading to great
instability and risk when using such system in real applications with
stringent safety requirement. Therefore, how to improve the robustness
of pattern recognition models in open world is a challenging and
important issue. The goal of this special issue is to broadly engage the
different communities together and provide a forum for the researchers
and practitioners related to this rapidly developed field to share their
novel and original research.

 

References:

[1]    W. J. Scheirer, A. de Rezende Rocha, A. Sapkota, and T. E. Boult,
Toward open set recognition, /IEEE Trans. Pattern Analysis and Machine
Intelligence/, vol. 35, no. 7, pp. 1757–1772, July 2013.

[2]    B. Biggio, F. Roli, Wild patterns: Ten years after the rise of
adversarial machine learning, /Pattern Recognition/, vol. 84, pp.
317-331, December 2018.

[3]    X.-Y. Zhang, C.-L. Liu, C. Y. Suen, Towards robust pattern
recognition: A review, /Proceedings of the IEEE/, vol. 108, no. 6, pp.
894-922, June 2020.

 

*Main Topics of Interest*

The special issue seeks original contributions which address the
challenges in open world robust pattern recognition. Possible topics
include but are not limited to:

Ÿ   Theoretical analysis of openness and robustness

Ÿ   Out-of-distribution detection, open-set recognition, anomaly detection

Ÿ   Class-incremental learning, continual learning, lifelong learning

Ÿ   Adversarial attack and defense methods

Ÿ   Learning with noisy data in open world

Ÿ   Model adaptation and transfer learning in changing environment

Ÿ   Applications in open world sensing such as video surveillance, robot
vision, autonomous driving, biometrics recognition, document analysis, etc.

 

*Submission Guidelines*

Papers should be formatted in a single column with double spacing and be
no more than 35 pages in length. The manuscript should be submitted via
the official website https://www.editorialmanager.com/pr/default.aspx
<https://www.editorialmanager.com/pr/default.aspx>. If you are not sure
whether your work is suitable to the special issue, please feel free to
contact the guest editors before the submission. To ensure that all
manuscripts are correctly identified for inclusion into the special
issue, it is important that authors select “SI: Open World Robust PR”
when they reach the “Article Type Name” step in the submission process.
We are happy to receive extensions of works presented in top conferences
but with a substantial revision (30 percent is generally considered
“substantial”).  Before submitting the manuscript, please read the
Instructions for Authors for Pattern Recognition journal
(https://www.elsevier.com/journals/pattern-recognition/0031-3203/guide-for-authors
<https://www.elsevier.com/journals/pattern-recognition/0031-3203/guide-for-authors>).
Authors should also explain the relevance of their submission to the
topic of the SI in their submitted paper, and how their submission
advances the state of the art in the topic of the SI.

 

*Important Dates*

Submission deadline: January 30, 2022

First review notification: April 20, 2022

Revised submission due: May 30, 2022

Final acceptance: August 1, 2022

 

*Guest Editors*

Ÿ   Xu-Yao Zhang, National Laboratory of Pattern Recognition, Institute
of Automation of Chinese Academy of Sciences. Email: xyz at nlpr.ia.ac.cn
<mailto:xyz at nlpr.ia.ac.cn>

Ÿ   Terrance E. Boult, Vision and Security Technology Lab, University of
Colorado Colorado Springs. Email: tboult at vast.uccs.edu
<mailto:tboult at vast.uccs.edu>

Ÿ   Cheng-Lin Liu, National Laboratory of Pattern Recognition, Institute
of Automation of Chinese Academy of Sciences. Email: liucl at nlpr.ia.ac.cn
<mailto:liucl at nlpr.ia.ac.cn>

Ÿ   Fabio Roli, Department of Electrical and Electronic Engineering,
University of Cagliari, Italy. Email: roli at diee.unica.it
<mailto:roli at diee.unica.it>

Ching Yee Suen, Centre for Pattern Recognition and Machine Intelligence,
Concordia University. Email: suen at cse.concordia.ca
<mailto:suen at cse.concordia.ca>

-- 
Terrance E. Boult,    Cell: (719)963-0573
El Pomar Prof. of Innovation and Security & Co-director Bachelor of Innovation 
U. Colorado at Colorado Springs
IEEE Fellow

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20210710/45744737/attachment-0001.html>