Call For Papers

The conference proceedings will be published by Springer in the Lecture Notes in Computer Science (LNCS) series. Award-winning papers will be invited to submit to a special issue of the International Journal of Computer Vision (IJCV).

Topics include (but are not limited to):

• Motion and Tracking
• Stereo and Structure from Motion
• Shape from X
• Low level vision
• Color and Texture
• Segmentation and Grouping
• Image-Based Modeling
• Deep Learning for Vision
• AI vocal remover
• Illumination and Reflectance Modeling
• Sensors & Early and Biologically-Inspired Vision
• Computational Photography and Video
• Object Recognition
• Object Detection and Categorization
• Vision and Language
• Video Analysis and Event Recognition
• Face and Gesture Analysis
• Statistical Methods and Learning
• Performance Evaluation
• Medical Image Analysis
• Document Analysis
• Optimization Methods
• RGBD and Depth Camera Processing
• Robotic vision
• Applications of Computer Vision

Submission Guidelines

• All submissions will be handled electronically via the conference’s CMT Website. Click the following link to go to the submission site: https://cmt3.research.microsoft.com/ACCV2018
• Example paper submission with detailed instructions: accv2018submission.pdf LaTeX Templates (zip): accv2018kit.zip
• A complete paper should be submitted using the above templates, which are blind-submission review-formatted templates.
• The submission page length is 14 pages for content plus maximum two pages for references. (max(main paper) = 14 pages; max(references) = 2 pages). Papers which violate these page limits will be rejected without review.

Poster Size

Posters should be printed A0 in size which is 84.1cm wide and 118.9cm tall.

The poster board size is 120cm wide and 180cm tall.

Instructions on Supplemental Material

Authors may optionally upload supplementary material (up to 100MB) which may include:

• videos to showcase results/demo of the proposed approach/system
• images and other results in addition to the ones in the paper
• anonymized related submissions to ACCV or other conferences and journals
• appendices or technical reports containing extended proofs and mathematical derivations that are not essential to the understanding of the submitted paper.

Only a single file with the either format of pdf, zip, mov or mp4 can be uploaded as supplemental material. Multiple files can be zipped into a single file and then be uploaded

ACCV encourages authors to submit videos using an MP4 codec such as DivX contained in an AVI. Also, please submit a README text file with each video specifying the exact codec used and a URL where the codec can be downloaded.

Authors should refer to the contents of the supplementary material appropriately in the paper. Note that reviewers will be encouraged to look at it, but are not required to do so. Please note that:

• All supplementary material must be self-contained in a single file for upload (i.e., a single pdf, or a zip file containing multiple items).
• The supplementary material directly supports the paper as submitted prior to the paper deadline. Only results generated by the algorithm/approach/system reported in the submitted version are allowed. Material based on improvements subsequent to the paper deadline is not allowed.
• DO NOT include links to external files or websites (e.g., youtube). Including such links is grounds for automatic rejection without review.
• DO NOT submit a newer version of the paper as supplementary material. Including a newer version of the paper or any portion thereof is forbidden.

Step-Wise Instructions For Online Paper Submission

Step 1: Create a user account on the CMT3 paper submission site. Skip this step if you already have a CMT3 account. To manage conflicts of interest during reviewer assignment, it is extremely important that each CMT3 user has one and only one CMT3 account for ACCV 2018, even if the user has multiple roles (e.g. author and area chair, author and reviewer). To ensure that each user has only one account, please let your co-authors know your email for ACCV 2018 for CMT3 purposes. If you do not yet have a CMT3 account:

• Follow this link to sign up as a new user on CMT.
• https://cmt3.research.microsoft.com/ACCV2018 Use this CMT3 link only; do not go to the old CMT site
• Make sure that your browser has cookies and Javascript enabled.
• Add “email@msr-cmt.org” to your list of safe senders (whitelist) to prevent important email announcements from being blocked by spam filters.
• Click on the ‘reset password’ button on the top right of the screen. Enter the code displayed and your CMT3 email address. Click submit. You will receive an email with your password.

Step 2: Enter conflict domains

• The first time you log on to CMT3 you will be asked to enter your “Conflict Domains”.
• Please enter the domain of the academic department or institution where you currently work or study (example: cs.toronto.edu), and also all institutions where you worked or studied or with which you have a very close collaboration in the last three years.
• DO NOT enter the domain of your email provider such as gmail.com, yahoo.com, hotmail.com, 163.com as your institution domain.

For each paper you want to submit, do the following:

Step 3: Enter a title
Click on the button “Create a new submission”. You will then be asked for the paper title. You can edit the title at any time up to the submission deadline.

Step 4: Add your co-authors
If your submission has co-authors, you will need to associate them with your submission.  For each co-author, enter the email address that corresponds exactly to their CMT3 account name. It is important that each person uses exactly one account as author, reviewer, or area chair for ACCV 2018. This will ensure that they can see your submission when they log onto CMT and allow us to handle conflicts appropriately.

Step 5: Select subject areas for your paper
Select one primary and up to four secondary.

Step 6: Upload your paper’s PDF (100MB limit)
Paper formatting instructions can be found above.

Step 7: Authorize the use of TPMS
Click on the ‘I agree’ checkbox

Step 8: submit
Click the submit button

Policies

Review Process: By submitting a paper to ACCV, the authors agree to the review process and understand that papers are processed by the Toronto system to help match each manuscript to the best possible chairs and reviewers.

Confidentiality: The review process of ACCV is confidential. Reviewers are volunteers, not part of the ACCV organisation and their efforts are greatly appreciated. The standard practice of keeping all information confidential during the review is part of the standard communication to all reviewers. Misuse of confidential information is a severe professional failure and appropriate measures will be taken when brought to the attention of ACCV organizers. It should be noted, however, that the organisation of ACCV is not and cannot be held responsible for the consequences when reviewers break confidentiality.

Double blind review: ACCV reviewing is double blind, in that authors do not know the names of the area chair/reviewers of their papers, and the area chairs/reviewers cannot, beyond reasonable doubt, infer the names of the authors from the submission and the additional material. Avoid providing information that may identify the authors in the acknowledgments (e.g., co-workers and grant IDs) and in the supplemental material (e.g., titles in the movies, or attached papers). Avoid providing links to websites that identify the authors. Violation of any of these guidelines may lead to rejection without review. If you need to cite a different paper of yours that is being submitted concurrently to ACCV, the authors should (1) cite these papers, (2) argue in the body of your paper why your ACCV paper is non trivially different from these concurrent submissions, and (3) include anonymized versions of those papers in the supplemental material.

Note that we will be actively checking for plagiarism.

Dual/Double Submissions: By submitting a manuscript to ACCV, authors acknowledge that it has not been previously published or accepted for publication in substantially similar form in any peer-reviewed venue including journal, conference or workshop. Furthermore, no paper substantially similar in content has been or will be submitted to a journal, another conference or workshop during the review period. The authors also attest that they did not submit substantially similar submissions to ACCV 2018. Violation of any of these conditions will lead to rejection. The goals of the dual submission policy are (i) to have exciting new work be published for the first time at ACCV, and (ii) to avoid duplicating the effort of reviewers. Our policy is based upon the following particular definition of “publication”. A publication, for the purposes of the dual submission policy, is defined to be a written work longer than four pages that was submitted for review by peers for either acceptance or rejection, and, after review, was accepted. In particular, this definition of publication does not depend upon whether such an accepted written work appears in a formal proceedings or whether the organizers declare that such work “counts as a publication”. Note that such a definition does not consider an arXiv.org paper as a publication because it cannot be rejected. It also excludes university technical reports which are typically not peer reviewed. However, this definition of publication does include peer-reviewed workshop papers, even if they do not appear in a proceedings, if their length is more than 4 pages including citations. Given this definition, any submission to ACCV should not have substantial overlap with prior publications or other concurrent submissions. As a rule of thumb, the ACCV submission should contain no more than 20 percent of material from previous publications. If you are making a submission to another conference or ACCV2018 at the same time, which covers similar or overlapping material, you should refer to that submission in order to explain the differences, just as you would if you had previously published related work. Authors are encouraged to contact the Program Chairs about clarifications on borderline cases. Note that a Technical Report (departmental, arXiv.org, etc.) version of the submission that is put up without any form of direct peer-review is NOT considered prior art and should NOT be cited. Likewise, mentioning a work under review in a presentation is NOT considered a violation.

Please note that the acceptance of the paper to the ACCV requires that at least one of the authors registers for the conference and present the paper there. Accepted papers will be published by Springer (with appropriate copyrights) electronically up to three weeks prior to the main conference. Please make sure to discuss this issue with your legal advisors as it pertains to public disclosure of the contents of the papers submitted.

Author Rebuttal Information

The deadline for submitting author rebuttals is September 7, 11:59pm Pacific time.

The rebuttal document must comply with the requirements below;

• It must be in PDF.
• It must use the ACCV submission format.
• It must be at most two pages long, including references.
• Hyperlinks to external files, websites, etc. are not allowed under any circumstances.

The reviewers and ACs are instructed to ignore rebuttals that do not obey these requirements.

Camera Ready Instructions

Author Kit

Please use accv2018CR_kit.zip as the template for the camera ready version of your paper.

More information about the camera ready guideline can be found here:

• Springer LNCS author information
• LNCS author guidelines
• LNCS LaTeX template

Copyright Forms

Authors must submit a signed Consent to Publish form, through which the copyright of their paper is transferred to Springer. Please download the template of the form here: ACCV Copyright form

Please fill this form, sign it, and scan it as a single PDF file, named XXXX-copyright.PDF, where XXXX is the four-digit paper ID (zero-padded if necessary). For example, if your paper ID is 24, the filename must be 0024-copyright.pdf. This file needs to be among the source files you are going to submit.

You will see that the conference name and the names of the volume editors are entered in advance. Please note that the main conference proceedings and the workshop proceedings have different copyright forms, and make sure you use the correct one. One author may sign on behalf of all of the authors of a particular paper, provided that permission to do so has been accorded by the other authors in advance. We do not accept digital signatures. If you have any queries regarding copyright, please contact Springer or the Publication Chair well in advance of publication. Accepted papers will be published by Springer (with appropriate copyrights) electronically after the main conference. Please make sure to discuss this issue with your legal advisors as it pertains to public disclosure of the contents of the papers submitted, and contact the Publication Chair if there is any sensitivity on the date of publication.

For ACCV Workshop papers, use this template instead:
ACCV Workshops Copyright form

Deadline

The submission deadline for camera ready material and copyright forms is 1 November, 2018. This deadline will not be extended. The ACCV 2018 proceedings will be published by Springer, in the LNCS series (Lecture Notes in Computer Science).

For each paper, at least one author has to be registered for attending the conference by the camera ready deadline. Without a valid registration at the conference by the deadline, your paper will not be published. Papers that arrive after the deadline may not appear on the conference LNCS proceedings and may not be placed in Springerlink.

Instructions and Policies

Paper length:

Papers may be up to 16 pages, maximum 14 pages for content plus maximum two pages for references. (max(main paper) = 14 pages; max(references) = 2 pages).

Author names:

Please write out author names in full in the paper, i.e. full given and family names. If any authors have names that can be parsed into FirstName LastName in multiple ways, please include the correct parsing in a comment to the editors, below the \author{} field.

How to submit camera ready papers:

We need all the source files (LaTeX files, style files, special fonts, figures, bib-files, Word or RTF files) that are required to compile papers, as well as the camera ready PDF. For each paper, one ZIP-file has to be prepared and submitted via the ACCV 2018 Submission Website, using the password you received with your initial registration on that site. The size of the ZIP-file may not exceed the limit of 100 MByte. The ZIP-file has to contain the following:

• All source files, e.g. LaTeX2e files for the text, PS/EPS or PDF/JPG files for all figures (RTF files for word-processing systems other than LaTeX/TeX).
  • Make sure to include any further style files and fonts you may have used.
  • References are to be supplied as BBL files to avoid omission of data while conversion from BIB to BBL.
  • Please do not send any older versions of papers. There should be one set of source files and one XXXX.pdf file per paper. Our typesetters require the author-created pdfs in order to check the proper representation of symbols, figures, etc.
  • You may use sub-directories.
  • Make sure to use relative paths for referencing files.
  • Make sure the source you submit compiles.
  • Please upload all the source files, including the camera ready XXXX.pdf, as a single ZIP file called XXXX.ZIP, where XXXX is the zero-padded, four-digit paper ID. Please upload this file (and optionally a supplemental file) in the “File Upload” page.
• PDF file named “XXXX.pdf” that has been produced by the submitted source, where XXXX is the four-digit paper ID (zero-padded if necessary). For example, if your paper ID is 24, the filename must be 0024.pdf. This PDF will be used as a reference and has to exactly match the output of the compilation.
• PDF file named “XXXX-copyright.PDF”: a scanned version of the signed copyright form (see above).

If you wish to provide supplementary material, the file name must be in the form XXXX-supp.pdf or XXXX-supp.zip, where XXXX is the zero-padded, four-digit paper ID as used in the previous step. Upload your supplemental file on the “File Upload” page as a single PDF or ZIP file of 100 MB in size or less.  Only PDF and ZIP files are allowed for supplementary material. You can put anything in this file – movies, code, additional results, accompanying technical reports–anything that may make your paper more useful to readers.  If your supplementary material includes video or image data, you are advised to use common codecs and file formats.  ACCV encourages authors to submit videos using an MP4 codec such as DivX contained in an AVI. Also, please submit a README text file with each video specifying the exact codec used and a URL where the codec can be downloaded. Authors should refer to the contents of the supplementary material appropriately in the paper.

Check that the upload of your file (or files) was successful either by matching the file length to that on your computer, or by using the download options that will appear after you have uploaded. Please ensure that you upload the correct camera-ready PDF–renamed to XXXX.pdf as described in the previous step as your camera-ready submission. Every year there is at least one author who accidentally submits the wrong PDF as their camera-ready submission.

For each paper, at least one author has to be registered for attending the conference by the camera ready deadline. Without a valid registration at the conference by the deadline, your paper will not be published. Both student and full registrations are fine for this purpose.

For any question related to the publication process, please do not hesitate to contact the publication chair.

Please kindly use the checklist below to deal with some of the most frequently encountered issues in ACCV submissions.

Files:

• My submission package contains ONE compiled pdf file for the camera-ready version to go on Springerlink.
• I have ensured that the submission package has all the additional files necessary for compiling the pdf on a standard LaTeX distribution.
• I have used the correct copyright form (with editor names pre-printed), and a signed pdf is included in the zip file with the correct file name.

Content:

• I have removed all \vspace and \hspace commands from my paper.
• I have not used \cite command in the abstract.
• I have read the Springer author guidelines, and complied with them, including the point on providing full information on editors and publishers for each reference in the paper (Author Guidelines – Section 2.8).
• I have entered a correct \titlerunning{} command and selected a meaningful short name for the paper.
• I have entered \index{Lastname, Firstname} commands for names that are longer than two words.
• I have used the same name spelling in all my papers accepted to ACCV and ACCV Workshops.
• I have not decreased the font size of any part of the paper to fit into 16 pages, I understand Springer editors will remove such commands.

Submission:

• All author names, titles, and contact author information is correctly entered in the submission site.
• The corresponding author e-mail is given.
• At least one author has registered by the camera ready deadline.

Call for Workshop Proposals

Workshops Chairs: Gustavo Carneiro (U Adelaide) and Shaodi You (ANU/Data61)
Proposal Deadline: 1 June 2018
Notification by: 11 June 2018

We are soliciting proposals for workshops to be held together with the 14th Asian Conference on Computer Vision (ACCV) 2018 in Perth, WA, Australia. Workshops will take place on the 2nd and 3rd of December, 2018.

The purpose of workshops is to provide a comprehensive forum on emerging topics that will not be fully explored in the main conference and to encourage in-­depth discussion of various technical and application issues in computer vision.  We also welcome “Challenge Workshops” that aim to compare new and established methods on common data sets.  Workshop registration, venue, as well as proceedings will be handled as part of the main conference by the ACCV organizers.

The selection process for workshop space, time and topic coverage will be competitive.  We will consider the topic coverage, the proposers’ credentials, the interested audience, and relevance of the topic in the selection process.  Please note that proposers may be asked to provide additional information, modify aspects of their proposals, or merge their proposal with another one.  Given that the camera-ready deadline is on 1st November, 2018, the workshop proposers must come with a submission/review schedule that meets this deadline.

Proposals should be submitted by email to the workshop chairs (accv18workshop@gmail.com) by 1st June, 2018.  The candidate workshop organizers will be notified of the decision by 11th June, 2018.

Proposals should be in PDF format and include the following information:

• Workshop title.
• Proposers’ names, titles, affiliations, and primary contact email.
• Topics that will be covered.
• Background and experience that makes the proposers well suited for organizing the workshop.
• Rough program outline (estimated numbers of orals, posters, and invited talks).
• Names and bios of any invited speakers and indication of whether they have agreed to speak.
• Anticipated target audience as well as expected number of attendees.
• Description of relevance and viability.
• Description of how this proposal relates to previous workshops at ACCV/CVPR/ICCV/ECCV/BMVC (be as specific as possible).
• Any special space or equipment requests.

For any questions, please contact the workshop chairs, Gustavo Carneiro (gustavo.carneiro@adelaide.edu.au) and Shaodi You (shaodi.you@data61.csiro.au).

Call for Tutorial Proposals

Tutorial Chair: Tat-Jun Chin
Last date to submit: 12 July 2018
Notification by: 26 July 2018

We are now open to receive proposals for short courses or tutorials at the 2018 Asian Conference on Computer Vision (ACCV 2018) in Perth. The tutorials will be held on 2nd December (Sunday) and 3rd December (Monday), prior to the main conference.

Tutorial topics should be of relevance and general interest to the computer vision community, and we encourage topics that are more established, and as well as newly emerging themes. A good tutorial is educational, in that it should

• Provide a comprehensive overview of a specific topic of interest in computer vision;
• Convey the fundamental theory or concepts underpinning the topic; and
• Summarise relevant experimental or technological milestones (if available).

The typical tutorial audience not only includes graduate or undergraduate students in computer vision, but also researchers and practitioners from academia and industry who are interested to learn about a new topic. Therefore, a good tutorial should also strive to be as self contained as possible.

We anticipate that each tutorial will be half-day (3 to 4 hours), unless the proposal justifies the need for additional time, e.g., the topic or tutorial speakers will be able to command significant and widespread interest.

We encourage tutorial proposers to refer to the accepted tutorials in previous ACCVs and other major computer vision conferences (CVPR, ICCV, ECCV):

www.accv2016.org/tutorial-page/
cvpr2018.thecvf.com/program/tutorials (also 2016 and 2017)
iccv2017.thecvf.com/program/tutorials
www.eccv2016.org/tutorials/

Proposals should be in PDF format and should include the following information:

• Title of tutorial;
• The names, titles, affiliations, emails, and bio sketches of the proposed tutorial speakers;
• If requesting full-day tutorial, justification for the additional time;
• Course description with list of topics to be covered, along with a brief outline and important details;
• Expected number of attendees;
• List of citations to relevant publications or other material by the organisers;
• A description of how this proposal relates to tutorials/short courses appearing at ACCV or other major computer vision conferences (CVPR, ICCV, and ECCV) within the last three years;
• Description of and/or links to any planned materials or resources to be distributed to attendees; and
• Indicate your preference for 2nd December or 3rd December to conduct your tutorial (we shall strive to include your preference, but kindly understand that it may not be possible to satisfy all constraints).

Proposals should be submitted by email to tat-jun.chin@adelaide.edu.au by 12 July 2018. Any questions should also be directed to the same email address.