AMV 2018

1st International Workshop on Advanced Machine Vision for Real-life and Industrially Relevant Applications

Monday, 3rd of December

A workshop in conjunction with ACCV 2018 in Perth, Australia.

Awards

At the AMV2018 workshop we arranged two awards. The best paper, accompagnied by a best paper award prize of 1000$ sponsored by iCetana, was assigned in the following way. From the top submissions (scored by reviewers and organizers) the best 4 papers were orally presented after which the audience could cast their vote for the best overall paper. Including those votes resulted in a tie between two papers. The workshop organizers then decided to assign the best paper award of the AVM2018 workshop to the paper that, in our opinion, would have the practical use for industrial applications. We are therefore proud to present that the best paper award went to Feras Almasri, for presenting his paper titled Multimodal Sensor Fusion In Single Thermal image Super-Resolution.

The best poster award was assigned in the following way. We asked all accepted submissions to prepare a poster for the poster session. We then allowed all visitors and the organizers of the poster session to vote for the best poster, which resulted in the best post award for the paper titled Markerless Augmented Advertising for Sports Videos by Emmanuel Cuevas and Iuliana Tabian.


             

Finally we would officially like to thank all the speakers for submitting papers to the AMV2018 workshop and all the attendees for making this workshop worth organizing. We clearly see there is a vast interested in a platform that brings together researchers and practitioners of advanced computer vision. As organizers we hope that the AMV workshop can be this platform for the following years to come.s We hope to see everyone at the AMV2019 workshop!


Home

We are proud to present the 1st International Workshop on Advanced Machine Vision for Real-life and Industrially Relevant Applications (AMV2018). This workshop is in conjunction with ACCV2018, Perth, Australia and is scheduled on Monday the 3rd of December 2018 (all-day workshop).

A large variety of industrially oriented applications (e.g. quality control, pick and place) have in the past decades been successfully implemented throughout a wide range of industries. These implementations are characterized by very controlled surroundings and objects (e.g. CAD models of objects available, controlled lighting). Advanced Machine Vision refers to computer vision - based systems where such assumptions do not hold, for example, when handling biological objects as seen in the food-production industry or when operating outdoors. With recent advancements in sensing and processing power, the potential for further automation in industry based on computer vision is clearly present. Furthermore, the exploding domain of computer vision algorithms (e.g. deep learning) provides dozens of new opportunities. However, there is in general a major gap between the topics in focus at major international computer vision conferences and the actual industrial needs. More often approaches are hardly transferable into practical and robust solutions for industrial challenges. The ambition of this workshop is to close this gap, by bringing together both academics and practitioners from the field.

The workshop will take place at the conference venue: The Perth Convention and Exhibition Centre (PCEC), 21 Mounts Bay Road, Perth, Australia.


Call For Papers

Please find a pdf version of this call for papers by clicking here

The ambition of this full-day AMV2018 workshop is to bring together practitioners and researchers from different disciplines related to Advanced Machine Vision to share ideas and methods on current and future use of computer vision algorithms in real-life and industrially relevant systems. This field raises the need of applied research that focusses on the technology transfer from academics towards practitioners, yielding several challenges like top-notch accuracies, real-time processing, minimal training data, minimal manual input, user-friendly interfaces, …

To this end we welcome contributions with a strong focus on (but not limited to) the following topics within Advanced Machine Vision:

  • Sensing

    • Camera selection
    • Camera setup
    • Different wavelengths
    • Multi-modal data
  • Improving robustness of algorithms

    • Real-time performance
    • Non-controlled illumination
    • Non-trivial intra object variability
    • Top-notch accuracies
  • Removing or reducing the need of training data

    • Data augmentation
    • Artificial data
  • Processing power and memory requirements

  • Obtaining training data and ground truth annotations

  • Lab testing versus inline testing

  • Transfer learning towards new applicational domains

  • Deep learning for advanced machine vision

  • Quality assessment of non-trivial objects

  • Real-life and industrially relevant applications

The workshop has a best paper award of $1000 sponsored by iCetana. On top of that there will be a special issue of the Machine Vision and Applications journal following the workshop. This special issue will be for both extended workshop papers as well as an open call.


Submission

Authors are encouraged to submit high-quality, original (i.e. not been previously published or accepted for publication in substantially similar form in any peer-reviewed venue including journal, conference or workshop) research.

The paper template is identical to the main ACCV2018 conference:

Papers are limited to 14 pages, excluding references. The review process is double blind. Papers that are not blind, or do not use the template, or have more than 14 pages (excluding references) will be rejected without review. For details concerning the blind review we refer to the example submission paper linked above. All workshop submissions follow the policies of ACCV 2018 (found here).

Submissions are handled through the CMT submission website: CLICK HERE FOR SUBMISSION

Authors are requested to submit their paper in a single PDF file (maximum file size 20MB). Submission of supplementary material is optional (up to 100MB). Accepted file formats for supplementary material are: PDF, PNG, JPG, GIF, ZIP, MP4, WMV, MPEG or AVI.

All accepted workshop papers will be published as workshop post-proceedings in the Lecture Notes in Computer Science (LNCS) by Springer, after the actual meeting, in order to exclude no-show papers from publication. On top of that, questions and feedback raised at the workshop can be used on top of the review comments to improve the camera-ready submission.

On top of that there will be a special issue of the Machine Vision and Applications journal following the workshop. This special issue will be for both extended workshop papers as well as an open call.

For questions/remarks regarding the submission e-mail: amv2018workshop@gmail.com.


Important Dates

  • EXTENDED paper submission deadline: 30th of September 2018 (11:59 CEST)
  • Notification of acceptance + reviews available: 26th of October 2018
  • Workshop date: Monday 3th of December 2018 - all day
  • Post-proceedings camera-ready deadline: TBD

People

Workshop Organizers


My profile picture.
Thomas B. Moeslund
Professor
Visual Analysis of People Lab
Aalborg University
Denmark

Website: click here
My profile picture.
Rikke Gade
Assistant Professor
Visual Analysis of People Lab
Aalborg University
Denmark

Website: click here
My profile picture.
Toon Goedeme
Professor
EAVISE Research Group
KU Leuven
Belgium

Website: click here
My profile picture.
Steven Puttemans
Post-doctoral researcher
EAVISE Research Group
KU Leuven
Belgium

Website: click here
My profile picture.
Ajmal Mian
Professor
School of Physics, Mathematics & Computing, The University of Western Australia
Australia

Website: click here

Program Committee


Benhur Ortiz Jaramillo, UGent
Bjorn Barz, Friedrich-Schiller-University Jena
Brian Booth, University of Antwerp
Christian Reimers, Friedrich-Schiller-University Jena
Clemens Brust, Friedrich-Schiller-University Jena
Dries Hulens, KU Leuven
Enrique Sanchez Lozano, University of Nottingham
Feng Chen, University of Nottingham
Isma Hadji, York University
Jie Chen, University of Oulu
Joachim Denzler, Friedrich-Schiller-University Jena
Kristof Van Beeck, KU Leuven
Lei Shi, University of Antwerp
Ljubomir Jovanov, UGent
Martin Kampel, Vienna University of Technology
Miguel Bordallo Lopez, University of Oulu
Mohammad Malekzadeh, Queen Mary University of London
Nikola Banic, University of Zagreb
Roxane Licandro, Medical University of Vienna, TU Wien
Sebastian Zambanini, TU Wien
Steve Vanlanduit, University of Antwerp
Tamas Suveges, University of Dundee
Thorsten Falk, University Of Freiburg
Timothy Callemein, KU Leuven
Tomislav Petkovic, University of Zagreb
Ulrich Schwanecke, Hoghschule RheinMain
Violeta Teodora Trifunov, Friedrich-Schiller-University Jena
Wiebe Van Ranst, KU Leuven
Xiaohua Huang, University of Oulu
Xiatian Zhu, Queen Mary University of London
Xin Liu, University of Oulu

The organizing committee would like to thank all members of the program committee for the work they invest in assuring that our AMV2018 workshop achieves a high-quality standard!


Program

Invited speakers

1) Yongsheng Gao

Vision Perception for Automation in Agriculture and Aquaculture

Environmental informatics studies new knowledge, technologies and devices for automation in agriculture and aquaculture, early detection of pest and plant disease, automatic species identification, plant phenomics, better water resource management, land environment monitoring, costal environment monitoring, marine life surveillance, etc. In this talk, he will introduce some of their work on automation in agriculture and aquaculture, faster grading and packing, species and cultivar identification, pest and disease recognition at Environmental Informatics @ Griffith, including recognition without detection, large image database retrieval (speed vs accuracy), and pose difference.

Biography

Professor Yongsheng Gao is the Director of Australian Research Council (ARC) Research Hub for Driving Farming Productivity and Disease Prevention, the founding Leader of the Environmental Informatics flagship group and the Director of Computer Vision and Image Processing Research Lab at Griffith University. He served as the project leader of Biosecurity Group, National ICT Australia (ARC Centre of Excellence) from 2009 to 2011. He is a current member of College of Experts (Panel Member), Australian Research Council. As a Chief Investigator, he has been working on projects in Australia, Singapore, Germany, and China in the areas of smart farming, biosecurity, face recognition, biometrics, image retrieval, computer vision, pattern recognition, environmental informatics, and medical imaging. He was also employed as a consultant by Panasonic Singapore Laboratories Pte Ltd working on the face recognition standard in MPEG-7. His research are reported in the media in Australia and Singapore, including The Australian, The Courier Mail, The Sydney Morning Herald, and The Straits Times (Singapore).

2) Faisal Shafait

PackageX: A journey from Deep Learning to Hand Crafting

Document Image Analysis and Recognition has served as a proven test bed for machine learning and computer vision research. Deep learning architectures like Convolutional Neural Networks and Long Short-Term Memory Networks made their mark on various document recognition problems before they got the attention of the broader research community. In this talk, I will present our work on reading package labels and how that evolved into PackageX, the last yard AI solution for managing incoming deliveries which is currently deployed across four continents. I will illustrate our journey that started with deep learning at its core but soon started heading towards classical pattern recognition techniques. In doing so, I will highlight various practical aspects that are often ignored in deep learning research but are crucial for deploying large scale systems in the real-world. I will conclude by providing some recommendations for design considerations when developing practical computer vision and machine learning systems.

Biography

Dr. Faisal Shafait is currently working as the Director of Deep Learning Laboratory at the National Center of Artificial Intelligence, Islamabad, Pakistan as well as a Professor at School of Electrical Engineering and Computer Science, National University of Sciences and Technology (NUST), Islamabad, Pakistan. Previously, he was an Assistant Research Professor at the School of Computer Science and Software Engineering at The University of Western Australia in Perth, Australia; a Senior Researcher at the German Research Center for Artificial Intelligence (DFKI), Kaiserslautern, Germany; and a Visiting Researcher at Google Inc., Mountain View, California. He received his PhD with the highest distinction in computer engineering from TU Kaiserslautern, Germany in 2008. His research interests include machine learning and pattern recognition with a special emphasis on applications in document image analysis. He has co-authored over 150 publications in international peer-reviewed conferences and journals in this area. He is serving as the Founding President of Pakistani Pattern Recognition Society, which is IAPR’s official chapter in Pakistan. Besides academics, Faisal has also been actively involved in a number of startup and spin-off companies and is currently working as CTO of VisionX Technologies, LLC.

Tentative program

Time What Who Title
08:30 Conference Registration Open - - - - - - -
09:00 Welcome - - - - - -
09:15 Keynote YongSheng Gao Vision Perception for Automation in Agriculture and Aquaculture
10:00 Coffee Break - - - - - -
10:30 Oral Paper 1 Sina Shafaei Integration of Driver Behavior into Emotion Recognition Systems: A Preliminary Study on Steering Wheel and Vehicle Acceleration
10:55 Oral Paper 2 Feras Almasri Multimodal Sensor Fusion In Single Thermal image Super-Resolution
11:20 Oral Paper 3 Zoltan Kato Simultaneous Multi-View Relative Pose Estimation and 3D Reconstruction from Planar Regions
11:45 Oral Paper 4 Quan Kong Multimodal Deep Neural Networks based Ensemble Learning for X-ray Object Recognition
12:10 Lunch Break - - - - - -
13:35 Keynote Faisal Shafait PackageX: A journey from Deep Learning to Hand Crafting
14:20 Oral Paper 5 Christian Bartz LoANs: Weakly Supervised Object Detection with Localizer Assessor Networks
14:45 Oral Paper 6 Yasir Jan WNet: Joint multiple head detection and head pose estimation from a spectator crowd image
15:10 Coffee break + Poster session - - - - - -
17:10 Best paper award - - - - - -
17:20 Closing remarks - - - - - -

Papers presented solely during poster session:

  • Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera
  • Unconstrained Iris Segmentation using Convolutional Neural Networks
  • Visual Siamese Clustering for Cosmetic Product Recommendation
  • PCA-RECT: An Energy-efficient Object Detection Approach for Event Cameras
  • Anomaly Detection Using GANs for Visual Inspection in Noisy Training Data
  • Prediction based deep autoencoding model for anomaly detection
  • Markerless Augmented Advertising for Sports Videos

All oral presentations will also have a poster available to foster further discussion on the topics presented.