AUTHOR GUIDELINES

Submissions Website
https://cmt3.research.microsoft.com/CVPR2022

Call for Papers

The call for papers in PDF format can be found here.

Important dates:

* Paper registration and submission dates are fixed, no extension will be given. 

Updates

• 10/20: Clarified social media policy; added FAQs on social media policy.
• 11/8: Clarified policy on authorship changes; added FAQs on authorship changes, changes to the paper details, and the questionnaire.
• 11/8: Added FAQ on the use of withdrawn datasets.
• 11/9: Clarified policy on IRB approval for Personal data/Human subjects
• 11/18: Clarified formatting of supplemental material. Clarified linking to large datasets in submitted code.
• 11/22: Clarified linking to additional image or video material.
• 1/24: Added description of text boxes on the rebuttal form

Policies

Important: Please note that policies have been updated significantly from the 2021 version and supplemented with ethics guidelines. Authors should carefully review the revised author guidelines as well as the ethics guidelines. Important changes include:

• Policy for data contributions
• Guidelines for the proper attribution of data assets
• Policy for the use of personal data or involvement of human subjects
• Guidelines for discussing potential negative societal impact
• Guidelines for discussing limitations

While conferences in adjacent academic communities, such as NeurIPS, have spent significant effort to introduce policies and guidelines for these important issues, and dealt with associated controversies, this is the first time CVPR is attempting to address such questions in a more concerted fashion. This year the intent is to increase awareness in the community and acquire experience for authors, reviewers, and organisers as well as to provide a basis for a discussion on how to handle such questions in future conferences. Note that many of these policies and guidelines are directly inspired by adjacent academic communities such as NeurIPS.

Review Process: By submitting a paper to CVPR, the authors agree to the review process and understand that papers are processed by the Toronto Paper Matching System (TPMS) to match each manuscript to the best possible area chairs and reviewers.

Confidentiality: The review process of CVPR is confidential. Reviewers are volunteers; they are not part of the CVPR organization and their efforts are greatly appreciated. The 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 the CVPR organizers. It should be noted, however, that the organization of CVPR is not and cannot be held responsible for the consequences when reviewers break confidentiality.

Conflict Responsibilities: It is the primary author's responsibility to ensure that all authors on their paper have registered their institutional conflicts into the submission system – CMT3 (see details under Domain Conflicts below). If a paper is found to have an undeclared or incorrect institutional conflict, the paper may be summarily rejected. To avoid undeclared conflicts, the author list is considered to be final after the submission deadline and no changes are allowed for accepted papers.

Double blind review: CVPR 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. Do not provide 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). Also do not provide 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 CVPR or another venue, the authors should (1) cite these papers; (2) argue in the body of your paper why your CVPR paper is non-trivially different from these concurrent submissions; and (3) include anonymized versions of those papers in the supplemental material.

Plagiarism: Plagiarism consists of appropriating the words or results of another, without credit. CVPR 2022's policy on plagiarism is to refer suspected cases to the IEEE Intellectual Property office, which has an established mechanism for dealing with plagiarism and wide powers of excluding offending authors from future conferences and from IEEE journals. You can find information on this office, their procedures, and their definitions of five levels of plagiarism at this webpage. We will be actively checking for plagiarism. Furthermore, the paper matching system is quite accurate. As a result, it regularly happens that a paper containing plagiarized material goes to a reviewer from whom material was plagiarized; experience shows that such reviewers pursue plagiarism cases enthusiastically.

Dual/Double Submissions: The goals of CVPR are to publish exciting new work for the first time and to avoid duplicating the effort of reviewers. By submitting a manuscript to CVPR, the 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, or archival forum. Furthermore, no publication substantially similar in content has been or will be submitted to this or another conference, workshop, or journal during the review period. Violation of any of these conditions will lead to rejection, and will be reported to the other venue to which the submission was sent.

A publication, for the purposes of this policy, is defined to be a written work longer than four pages (excluding references) 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”.

As per PAMI TC motion, the above definition does not consider an arXiv.org pre-print 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 four pages (excluding citations). Given this definition, any submission to CVPR should not have substantial overlap with prior publications or other concurrent submissions.

A submission with substantial overlap is one that shares 20 percent or more material with previous or concurrently submitted publications. 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 in the submission.

Data contributions: If a paper submission is claiming a dataset release as one of its contributions, it is expected that the dataset will be made publicly available no later than the camera-ready deadline. To facilitate handling papers with dataset contributions, the authors will need to indicate in CMT at submission time whether the submitted paper claims a dataset as one of its main scientific contributions. This information will be visible to the reviewers and area chairs, who will be asked to verify during the review process whether the authors’ indication is consistent with the paper content. Any paper that claims a dataset contribution will need to provide a URL for the dataset when submitting the camera ready. Once the online proceedings are released, the CVPR website will provide a list of all papers with dataset contributions and include a link to the author-provided URL in order to facilitate wide dissemination of new datasets to the CVPR audience.

Note that this does NOT imply that all datasets used in CVPR submissions must be public. The use of private or otherwise restricted datasets (e.g. for training or experimentation) continues to be permitted. However, private or otherwise restricted datasets cannot be claimed as contributions of the paper as they do not become available to the scientific community.

Use of existing assets: Just like papers are expected to cite previous work that inspired a submission or on which a submission is built, we expect CVPR papers to cite assets, such as code or datasets, that have been used in the creation of the submitted manuscript. If there are multiple versions of an asset, specify the version you have been using. This attribution of assets can be made either in the main paper or in the supplemental material. We furthermore encourage authors to discuss the license and/or copyright terms of the assets used. The inclusion of a URL is encouraged as well, where appropriate.

Personal data / human subjects: If a paper makes use of personal data and/or data from human subjects, including personally identifiable information or offensive content, we expect that the collection and use of such data has been conducted carefully in congruence with the ethics guidelines. In many countries and institutions, the collection and use of personally identifiable data or data from human subjects is subject to approval from an Institutional Review Board (IRB, or equivalent). If the use of such data was approved by an IRB, stating this is sufficient. If the use of such data has not (yet) been approved by an IRB, authors should provide information on any pending approval process, how the data was obtained, as well as discuss if and how consent was obtained (or why it, perhaps, could not be obtained). This discussion can be included either in the main paper or in the supplemental material.

IRB reviews for the US or the appropriate local ethics approvals are typically required for new datasets in most countries. It is the dataset creators' responsibility to obtain them. If the authors use an existing, published dataset, we encourage, but do not require them to check how data was collected and whether consent was obtained. Our goal is to raise awareness of possible issues that might be ingrained in our community. Thus we would like to encourage dataset creators to provide this information to the public.

Discussion of potential negative societal impact: As the AI community at large has been paying increasing attention to the issue of negative societal impact of research, CVPR 2022 aims to raise awareness of potential negative societal impact in the CVPR community as well, as described in the ethics guidelines. To that end, we encourage all authors to think about this issue in the context of the technologies they developed and discuss that in their paper. Note that there is no formal requirement to include a discussion of potential negative societal impact, but reviewers will be asked to favorably consider the inclusion of a meaningful discussion of this issue.

Examples of negative societal impact include potential malicious or unintended uses (e.g., disinformation, generating fake profiles, surveillance), environmental impact (e.g., training huge models), fairness considerations (e.g., deployment of technologies that could further disadvantage historically disadvantaged groups), privacy considerations (e.g., a paper on model/data stealing), and security considerations (e.g., adversarial attacks).

We expect many papers to be foundational research and not tied to particular applications, let alone deployments. But being foundational does not imply that research has no negative societal impact. If authors see a direct path to any negative applications, they should point it out, even if it is not just specific to their work. In a theoretical paper on algorithmic fairness, the paper might caution against overreliance on mathematical metrics for quantifying fairness and examples of ways this can go wrong. If the paper improves the quality of generative models, authors might point out that their approach can be used to generate deep-fakes for disinformation. On the other hand, if the paper develops a generic algorithm for optimizing neural networks, authors do not need to mention that this could enable people to train models that generate deep-fakes faster.

Such discussion shall consider different stakeholders that could be impacted by the authors’ work. It is possible that research benefits some stakeholders while harming others. We encourage the discussion to pay special attention to vulnerable or marginalized communities.

Such discussion shall also consider possible harms that could arise when the technology is being used as intended and functioning correctly, harms that could arise when the technology is being used as intended but gives incorrect results, and harms following from (intentional or unintentional) misuse of the technology.

If there is negative societal impact, the discussion should also include any mitigation strategies (e.g., gated release of models, providing defenses in addition to attacks, mechanisms for monitoring misuse, mechanisms to monitor how a system learns from feedback over time, improving the efficiency and accessibility of CV models).

The applications of a technology often could have two sides. A thorough discussion of potential societal impact is beneficial as it raises awareness in the community. Authors should not be concerned that including a discussion could be viewed negatively. Reviewers and area chairs will be asked to favorably consider the inclusion of a meaningful discussion of potential negative societal impact. While in rare cases, papers with glaring negative consequences might be seen critically by the reviewers, this is independent of whether this negative impact is discussed in the paper or not. 

Discussion of Limitations: Considering the limitations of an approach is an important part of good academic scholarship. While many high-quality CVPR papers thoroughly consider and discuss limitations, CVPR 2022 aims to emphasize the importance of an honest treatment of limitations. Authors are thus explicitly encouraged to discuss limitations, ideally by creating a separate "Limitations" (sub-)section in their papers. 

Such discussion shall point out any strong assumptions and how robust the results are to violations of these assumptions (e.g., independence assumptions, noiseless settings, model well-specification, asymptotic approximations only held locally). Authors need to reflect on how these assumptions might be violated in practice and what the implications would be. 

The discussion shall also reflect on the scope of their claims, e.g., if the authors only tested their approach on a few datasets or did a few runs. In general, empirical results often depend on implicit assumptions, which should be articulated. The discussion shall reflect on the factors that influence the performance of the approach. For example, a recognition algorithm may perform poorly when image resolution is low or images are taken in low lighting.

We understand that authors might fear that complete honesty about limitations might be used by reviewers as grounds for rejection. It is worth keeping in mind that a worse outcome might be if reviewers discover limitations that are not acknowledged in the paper. In general, we advise authors to use their best judgement and recognize that individual actions in favor of transparency play an important role in developing norms that preserve the integrity of the community. Reviewers will be specifically instructed to not penalize honesty concerning limitations

Attendance responsibilities: The authors agree that if the paper is accepted, at least one of the authors will register for the conference and present the paper there.

Publication: All accepted papers will be made publicly available by the Computer Vision Foundation (CVF) two weeks before the conference. Authors wishing to submit a patent understand that the paper's official public disclosure is two weeks before the conference or whenever the authors make it publicly available, whichever is first. The conference considers papers confidential until published two weeks before the conference, but notes that multiple organizations will have access during the review and production processes, so those seeking patents should discuss filing dates with their IP council. The conference assumes no liability for early disclosures. More information about CVF is available at http://www.cv-foundation.org/.

Publicity, social media: Papers submitted to CVPR must not be discussed with the press until they have been officially accepted for publication. Submitted papers as well as work with substantial overlap to the CVPR submission (such as an arXiv paper) also may not be advertised on social media. Please see the FAQ section for more details. Violations may result in the paper being summarily rejected or removed from the conference and proceedings.

Social Media Ban: Per the motion passed in the CVPR 2021 PAMI-TC meeting, authors should NOT use social media to promote their paper submissions to CVPR during the review period. We are imposing a policy slightly stronger than what passed in the motion, where we define the social media silence period. By our definition, the social media silence period starts four weeks before the paper submission deadline, until the final paper decision notifications are sent to authors. Per the currently planned schedule, the social media silence period is from 10/19/2021 to 03/02/2022. Any social media promotion of a paper incurred in this period, proactively initiated by the authors, is considered a policy violation. 

Social media promotion of a paper submission not only runs a bigger risk of violating the double blind review policy, but also imposes an even stronger bias into the process, which may influence reviewers’ technical judgement of a paper

Authors acting as reviewers: Given the growth of the number of paper submissions, we expect all authors to be willing to serve as reviewers as well. With a large enough pool of reviewers, we expect that reviewers will be sent on the order of five papers to review. Our timeline expects that papers will go to reviewers about December 16, 2021, and reviews should be returned on or before January 14, 2022.

Submission Guidelines

All submissions will be handled electronically via the conference's CMT Website. By submitting a paper, the authors agree to the policies stipulated in this website. The paper submission deadline is November 16, 2021 (extended to November 18, 2021). Note that each paper should already be registered one week earlier, by November 9, 2021. Supplementary material can be submitted until November 23, 2021 (extended to November 24, 2021).

Papers are limited to eight pages, including figures and tables, in the CVPR style. Additional pages containing only cited references are allowed. Please refer to the following files for detailed formatting instructions:

• Example submission paper with detailed instructions PDF
• CVPR 2022 Author Kit for LaTeX/Word Zip file (updated 08/23/2021)

Papers that are not properly anonymized, or do not use the template, or have more than eight pages (excluding references) will be rejected without review.

1) Paper submission and review site:

Submission Site (bookmark or save this URL!)

Please make sure that your browser has cookies and Javascript enabled.

Please add "email@msr-cmt.org" to your list of safe senders (whitelist) to prevent important email announcements from being blocked by spam filters.

Log into CMT3 at https://cmt3.research.microsoft.com. If you do not see “IEEE/CVF Conference on Computer Vision and Pattern Recognition 2022” in the conference list already, click on the “All Conferences” tab and find it there.

2) Author registration: 

When you log in for the first time, you will be asked to set up a user profile and provide their conflict information. You will not be able to submit any paper without entering this information. At any time, you can update this information by clicking on your name in the upper-right and selecting “User information” under CVPR2022.

It is the primary author's responsibility to ensure that all authors on their paper have registered on CMT3 and provide their conflict information.

1. User Profile: Here you can update the basic information on your CMT profile.
    
2. Conflict Information: Authors need to provide information on conflict domains. Each author should list domains of all institutions they have worked for, or have had very close collaboration with, within the last 3 years (example: mit.edu; ox.ac.uk; microsoft.com). DO NOT enter the domain of email providers such as gmail.com. This institutional conflict information will be used in conjunction with prior authorship conflict information to resolve assignments to both reviewers and area chairs. If a paper is found to have an undeclared or incorrect institutional conflict, the paper may be summarily rejected.
    
3. Creating a Paper Submission: This step must be completed by the paper registration deadline. After this deadline, you will not be able to register new papers, but you will be able to edit the information for existing papers. While it will be possible to make minor edits to the title and abstract until the full paper submission deadline, submissions with “placeholder” abstracts that are rewritten entirely for the full submission may be removed without consideration. Note: we are allowing changes to the author list until the full paper deadline. After that, no changes will be permitted for any reason, including for the camera-ready version.
   
   (a) Click the “+ Create new submission” button in the upper-left to create a new submission. There, you will be prompted to enter the title, abstract, authors, and subject areas. You are strongly encouraged to finalize the author list by the registration deadline.
   
   (b) Check with your co-authors to make sure that: (1) you add them with their correct CMT3 email; and (2) they have logged in to the submission website and filled out the conflict information on CMT3. If you add an author with an email that is not in CMT3 and the name and organization is not automatically filled, that means they are not yet in the system, and you should make sure to check that they do not already have an account under a different email before completing the requested information to add them.
   
   (c) Enter subject (topic) areas for your paper. You must include at least one primary area and up to 10 secondary areas. This information is used to help assign ACs and reviewers.
    
4. Paper Number: Once you have registered your paper (i.e. title/authors), you will be assigned a paper number. Insert this into the LaTeX or Word template before generating the PDF of your paper for submission. Papers submitted without a number may not be reviewed.
    
5. Authorship Changes:
   After the paper submission deadline, the list of authors will be considered final. After that date, new authors CANNOT be added; authors may also NOT be removed. Changes to the authorship order will be permitted only in exceptional circumstances.
    
6. Submission Requirements:
   * The maximum size of the abstract is 4000 characters.
   * The paper must be PDF only (maximum 50MB). Make sure your paper meets the formatting and anonymity requirements described above.
   * The supplementary material can be either PDF or ZIP only (maximum 100MB).
    
7. Supplementary Material Submission: By the supplementary material deadline, the authors may optionally submit code and/or additional material that was ready at the time of paper submission but could not be included due to constraints of format or space. The authors should refer to the contents of the supplementary material appropriately in the paper. Reviewers will be encouraged to look at it, but are not obligated to do so.

Supplementary material may include videos, proofs, additional figures or tables, more detailed analysis of experiments presented in the paper, code, or a concurrent submission to CVPR or another conference. It may not include results on additional datasets, results obtained with an improved version of the method (e.g., following additional parameter tuning or training), or an updated or corrected version of the submission PDF. Make sure the supplementary material does not reveal author identity. Papers with supplementary materials violating the guidelines may be summarily rejected.

8. Code Submission and Reproducibility: To improve reproducibility in AI research, we highly encourage authors to voluntarily submit their code as part of the supplementary material. Authors should also use the Reproducibility Checklist as a guide for writing reproducible papers. Reviewers are encouraged to check the submitted code to ensure that the paper’s results are trustworthy and reproducible. The code should be anonymized, e.g., author names and institutions (e.g. also in license / copyright statements) should be removed. The authors should include instructions for how to set up the environment and run the code. The code does not need to cover all experiments, but should aim to cover the main results and as many of the minor experiments as possible. We do not expect authors to submit private/sensitive data, only sufficient data to demonstrate the method. All code/data will be reviewed confidentially and kept private.

Detailed supplementary material guidelines: 

(a) All supplementary material must be self-contained and zipped into a single file. The following document and media formats are allowed: avi, mp4, pdf, wmv. CMT imposes a 100MB limit on the size of this file. Note that you can update the file by uploading a new one (after removing the previous version).

(b) The paper for review (PDF only) must be submitted first before the supplementary material (PDF or ZIP only) can be submitted. 

(c) There are no specific formatting requirements for PDFs in the supplementary material. The official CVPR style may be – but does not have to be – used. Legibility and clarity are always appreciated, especially as the reviewers are not obliged to look carefully.

(d) Code can be submitted as part of the supplementary ZIP file or through anonymous GitHub repositories (include the link in a separate text file in the supplementary ZIP). The link should point to a branch that will not be modified after the submission deadline.  Link or repository access should not reveal author or reviewer identity.

Rebuttal Instructions

After receiving the reviews, the authors may optionally submit a rebuttal to address the reviewers' comments, which will be limited to a one page PDF file using the CVPR 2022 Rebuttal Template.

The rebuttal must maintain anonymity and cannot include external links that reveal the author identity or circumvent the length restriction.

Responses longer than one page will simply not be reviewed. This includes responses where the margins and formatting are deemed to have been significantly altered from those specified by the style guide.

The author rebuttal is optional and is meant to provide you with an opportunity to rebut factual errors or to supply additional information requested by the reviewers. It is NOT intended to add new contributions (theorems, algorithms, experiments) that were absent in the original submission and NOT specifically requested by the reviewers. You may optionally add a figure, graph, or proof to your rebuttal to better illustrate your answer to the reviewers' comments.

Per a passed 2018 PAMI-TC motion, reviewers should refrain from requesting significant additional experiments for the rebuttal, or penalize for lack of additional experiments. Authors should refrain from including new experimental results in the rebuttal, especially when not specifically requested to do so by the reviewers.

New this year: Authors can respond to questions raised about ethics/transparency/privacy/fairness/social impact separately in a text box on the rebuttal form. This text box should only be used for this specific purpose – all other points, including questions about limitations, must be addressed in the rebuttal PDF. Any unrelated technical discussion in this box will simply be discounted.

Confidential comments: Authors have the possibility to submit confidential comments to the area chair (meta reviewer) in a text box on the rebuttal form. Please do so only in exceptional circumstances, e.g. if there are grave misunderstandings on the part of a referee.

Author FAQs

About Submitting Papers

Q. Can we please have an extension on the paper registration or submission deadline?

A. NO. And any incomplete submission or a submission not meeting required criteria will be deleted.

Q. Can we get my quota increased for the size of paper submission from 50 MB to something higher?

A. NO. We have set hard limits of 50MB (PDF Only) for paper submission and 100MB (PDF or ZIP only) for supplementary materials for submissions for review. As we are expecting many submissions, and as each reviewer is expected to review multiple papers, larger file downloads (and uploads) will tax the system and abilities of reviewers to get to the papers fast enough. Authors should consider adding hi-res images as supplementary material (see supplementary material guidelines).

Q. How do I delete Supplementary Material from the CMT site?

A. After you log in, in the "Author" console, you'll notice "Upload/Delete File" at the end of the supplementary file name. Click on that, and in the page that appears, you can click on the "Delete" button to remove the supplementary file. (Please note that you will not be able to delete the supplementary file after the supplementary material deadline.)

Q. Can we submit color images with our papers for review?

A. YES. Reviewers will get the exact PDF file of the paper you submitted, so they can see the color images on the screen. Do be warned though that many reviewers still like to read printed papers and not all have access to high-end color printers. Please make sure to comment in the paper to request the reviewers to see the color online copy. If you use color in your plots, please keep in mind that a significant subset of reviewers and readers may have a color vision deficiency; red-green blindness is the most frequent kind. Hence avoid relying only on color as the discriminative feature in plots (such as red vs. green lines), but add a second discriminative feature to ease disambiguation.

Q. Are there any formatting requirements for PDFs in the supplemental material?

A. No. The important thing is that supplementary PDFs are legible and neatly formatted. Many authors choose to use the official CVPR style for any supplementary PDFs as well, but this is not a must. Formatting supplementary documents in a single-column layout is permitted

Q. Can I link to additional image or video material from the supplemental material?

A. Only if absolutely necessary and as long as the review double-blind review process and deadline integrity are preserved. To that end, authors need to ensure following conditions: (1) The image and video material is too large to include in the 100MB file size limit. (2). The hosting site and the linked material does not reveal the identity and affiliation of the authors. (3) The hosting site or apps do not track and identify who viewed the materials. (4) The authors provide a smaller-sized version of their image or video material in the submitted supplemental material.
Condition 4 ensures that reviewers have a direct way of viewing the material (albeit at a lower quality) and are also able to verify that the externally hosted material has not been modified since the supplemental material deadline.
Authors bear the responsibility and are advised to proceed with caution not to break the double-blind review process. Note, not all hosting services are available in all regions. Authors should also note that, just like for the supplemental material itself, reviewers are under no obligation to review such additional image or video material.

Q. Can I add/remove authors after the paper deadline?

A. NO. After the paper deadline, the author list is considered final. Changes to the authorship order may be considered, but only in special circumstances.

Q. Can I update my paper’s information (e.g., title, abstract, author list, subject areas, questionnaire) after the paper registration deadline?

A. YES, you can update your paper’s information until the paper submission deadline. However, after the registration deadline, you can no longer create new paper submissions in CMT.

Q. Where do I find the questionnaire? Do all authors need to fill this out?

A. A questionnaire needs to be filled out when creating a paper submission (i.e. after clicking “Create new submission” in CMT). Only one author needs to provide this information. However, all authors need to provide their conflict domains

Q. What is CVPR 2022 policy on DUAL SUBMISSIONS?

A. Please read the dual/double submission paragraph above.

Q. Does a Technical Report (departmental, arXiv, etc.) available online count as a prior publication, and therefore is that work ineligible for review and publication at CVPR 2022?

A. Please read the dual/double submission paragraph above.

Q. Does a document on GitHub or other open repositories count as a publication, and therefore is ineligible for review and publication at CVPR 2022?

A. Submissions to GitHub and similar repositories cannot be rejected and are accepted by default before any "review" that can take place on such platforms. Given definitions in the dual/double submission paragraph above, GitHub documents are not publications and won't be treated as such. To preserve anonymity, you should not cite your public codebase. You can say that the code will be made publicly available.

Q. Does a presentation at a departmental seminar during the review period violate the anonymity standard or other CVPR 2022 policy?

A. NO. Authors must properly anonymize the written submission as per the guidelines. Presentation of material at an academic talk is acceptable.

Q. Can I promote my paper in the press or on social media?

A. As stated in the instructions above, the authors are not allowed to go to the press with their submission prior to the end of the review process, or to advertise their work orwork with substantial overlap to the CVPR submission (such as an arXiv paper) on social media. In recent conference cycles, some authors were found posting about their submissions on Twitter or other social media, even including the title or a snapshot of the paper. This is a violation of anonymity, since the message may go out to many potential reviewers. 

To avoid promotion right before the media ban, the social media silence period starts four weeks before the submission deadline and ends upon final author notification. This year this is projected to be from 10/19/2021 to 03/02/2022. 

Authors must not:

• Talk to the media about their work as "in submission to CVPR".
• Make any posts to social media or elsewhere that can be linked to a specific CVPR submission (e.g., mentioning the title of the submission or details and content). 

Authors may:

• Talk about their work in an academic presentation without saying it's submitted to CVPR.
• Submit to arXiv without mentioning CVPR.

A paper may be rejected if the program chairs feel that the authors have attempted to let potential reviewers know who wrote the paper.

Q. Can my post mention a paper that is in review (without mentioning CVPR)?

A. No. Per the motion passed at CVPR 2021, this is a loophole to the original press ban principles.

Q. If a social media post shared information of a CVPR submission without the authors being involved, does that constitute a violation?

A. No, it does not. A violation occurs only when authors are proactively doing so.

Q. May I have a video link in my arXiv paper?

A. Yes, you may and it is not considered a violation of the social media ban, as long as the video (or a link of the video) is not posted on any social media platform, and it does not contain any information that would otherwise link it to your CVPR submission. In addition, if your video is hosted on a video platform that behaves similar to a social media platform (such as YouTube, Douban, or Blibli) then your video post must be unlisted and have comments disabled. 

Q. May I build a project website related to my arXiv paper?

A. Yes, you may and it is not considered a violation of the social media ban, as long as the url of the project website is not posted on any social media platform, and the project website itself does not contain any information that would otherwise link it to your CVPR submission.

Q. Why is social media promotion an issue?

A. Groups with large followings and the resources to mount visible social media promotions can receive significant attention for work that is under review. Reviewers are exposed to this work and the attention it receives can bias their judgment - if so many people on social media are excited, mustn't it be good? Groups with fewer followers or that refrain from such campaigns are disadvantaged. This biases the peer review process and reduces trust in its fairness.

Peer review is the backbone of science. The process helps detect mistakes or false claims before the work appears in public. This reduces the chance that work needs to be retracted and, hence, increases public trust in science and the scientific process. Science depends on this trust both for funding and for its independence. Anything that undermines this trust can have long term negative consequences for basic research.

Q. What about arXiv papers?

A. The field has decided that dissemination on arXiv facilitates the rapid spread of information within the field. arXiv papers are not “published” but are understood to be "pre-publications." This open pre-publication process provides a form of community review where problems can be detected (much like formal peer review). arXiv papers are often corrected and modified; the site is set up to support this scientific process of revision. Putting a paper on arXiv for early analysis by experts is very different from publicly promoting work on social media to a broad audience. For example, attention to a particular arXiv paper cannot be gathered from arXiv itself.

Q. arXiv tweets new papers. Is that a violation?

A. No. This is an automatic process and does not constitute the authors promoting their work. arXiv tweets are largely followed by experts in the field and not the general public. The work is presented in its entirety and a pre-publication and can be judged as such. This differs from, for example, promotional videos posted on social media.

Q. Doesn’t this slow down scientific progress?

A. No. Experts in the field make scientific progress, not the general public. The exemption for arXiv means that the research community still gets early access to research and can evaluate it as non-peer-reviewed.

Q. Why the longer period of social media silence?

A. This is to discourage authors from trying to circumvent the ban by posting shortly before the submission deadline. The assumption is that four weeks before the deadline authors have a good idea of what they will be submitting to CVPR.

Q. Can I respond to someone else’s post about my arXiv submission?

A. You should err on the side of caution. If such a response makes it easy to establish your identity, then it will be considered a violation of anonymity and proactive engagement with social media on the paper. Posting pseudonymously, if it is clear that it is the author posting, could still be considered a proactive engagement with social media

Q. How do I cite my results reported in open challenges?

A. To conform with the double blind review policy, you can report results of other challenge participants together with your results in your paper. For your results, however, you should not identify yourself and should not mention your participation in the challenge. Instead present your results referring to the method proposed in your paper and draw conclusions based on the experimental comparison to other results.

Q. Does my submission need to cite arXiv papers that are related to my work?

A. Consistent with good academic practice, you need to cite all sources that inspired and informed your own work. This said, asking authors to thoroughly compare their work with arXiv reports that appeared shortly before the submission deadline imposes an unreasonable burden. We also do not wish to discourage the publication of similar ideas that have been developed independently and concurrently. Authors and reviewers should keep the following guidelines in mind:

• Authors are not required to discuss and compare their work with recent arXiv reports, although they must properly cite those that inspired them.
• To reduce confusion, whenever citing papers that initially appeared on arXiv, the authors should check whether those papers had subsequently been published in a peer-reviewed venue, and to cite those versions accordingly.
• Failing to cite an arXiv paper or failing to beat its performance SHOULD NOT be sole grounds for rejection.
• Reviewers SHOULD NOT reject a paper solely because another paper with a similar idea has already appeared on arXiv. If the reviewer suspects plagiarism or academic dishonesty, they are encouraged to bring these concerns to the attention of area and program chairs.
• It is acceptable for a reviewer to suggest that an author should acknowledge or be aware of something on arXiv.

About the Review Process

Q. Is the CVPR 2022 Review Process CONFIDENTIAL?

A. YES, CVPR 2022 Reviewing is considered confidential. All reviewers are required to keep every manuscript they review as confidential documents and not to share or distribute materials for any reason except to facilitate the reviewing of the submitted work.

Q. Are CVPR 2022 Reviews Double BLIND or Single BLIND?

A. CVPR reviewing is Double BLIND, in that authors do not know the names of the area chair/reviewers of their papers, and area chairs/reviewers do not know the names of the authors. Please read Section 1.7 of the example paper ReviewTemplate.pdf for detailed instructions on how to preserve anonymity. 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 will lead to rejection without review.

About the Data Contributions Policy

Q. Is releasing a dataset required.

A. NO, it is completely optional.

Q. Does the data contribution policy mean that any dataset that I use must be public?

A. NO, you may rely on private or otherwise restricted datasets. The policy applies only to those papers that claim the release of a dataset as one of its core scientific contributions. If you claim that the release of a dataset is one of your paper’s key contributions, then the expectation is that the dataset will be released. A link to the dataset shall be provided when submitting the camera-ready paper.

Q. I use a private dataset for my experiments that I cannot distribute publicly. Can I still submit a paper?

A. YES, you can absolutely submit a paper. You simply cannot claim the dataset as one of the paper’s contributions. The paper must stand on its other scientific merits (e.g. its technical contribution).

Q. I wish to claim a dataset contribution in my paper but I cannot release the dataset. Is this possible?

A. NO, this is not consistent with the dataset policy. If the release of a dataset is one of your key scientific contributions, it is the expectation that the data is made available to the research community. Conversely, if releasing the dataset is not possible, then it should not be a key contribution of your paper. In such a case your paper must stand on its other (e.g. technical) contributions.

Q. I wish to claim a dataset contribution in my paper and release the dataset, but the data is not ready for release at submission time. Is this a problem?

A. NO, you can make minor modifications to the dataset until the camera-ready deadline. However, the scientific conclusions based on the dataset at submission time must be sound and must continue to hold. You must provide a link to the dataset only when submitting the camera ready. 

Q. I wish to claim a dataset contribution in my paper and release the dataset, but the data may/will not be ready for release at the camera-ready deadline. Is this an issue?

A. YES, it is the expectation that your dataset will be ready and available at the time when submitting the camera ready, since at this point the paper that describes the dataset is considered final as well. If you cannot ensure that you can meet this deadline, then the release of the dataset should not be one of the major scientific contributions of your paper.

Q. I plan to release a dataset with my paper, but my dataset website requires users to create a log-in or otherwise request permission, before they can access the dataset. Is this permitted?

A. YES, this is permitted as long as you do not unreasonably withhold or delay access to members of the research community.

Q. Releasing the dataset is not fully under my control, e.g. because it needs to still be approved by an institutional review board. What should I do?

A. We recommend waiting with your submission until you have obtained the appropriate approval.

Q. Are there any specific license requirements for dataset contributions?

A. NO, the expectation is that the data is available for research use by members of the research community.

 

About the Attribution of Assets (Datasets & Code)

Q. Do I need to attribute assets that I used in my paper?

A. YES, you should treat assets such as datasets or code just like scientific papers.

Q. Do I really need to cite ALL assets that I have been using? For example, do I need to include a citation to Python, since I used Python to write my code?

A. NO, you should cite research assets. Use good judgement, just like you would for citations of previous scientific work. As an example, most papers will happily make use of the Gaussian distribution without making reference to Carl Friedrich Gauss. Similarly, it is fine to use general purpose programming languages or widely used programming tools without explicit citation. Research datasets or specific algorithms, even if widely used such as ImageNet or Adam, call for a citation, on the other hand. 

Q. Do I need to include URLs of assets?

A. You are encouraged, but not required to do so. Including a URL is particularly sensible if the asset is not widely used or if the URL is not easily associated with the paper that introduced the asset.

Q. Do I need to discuss the licensing terms / copyright terms of the assets?

A. You are not required but encouraged to do so. It should go without saying that assets should be used in accordance with their license / copyright terms. Including information on the licensing terms / copyright terms shows that authors carefully considered this. 

 

About Personally Identifiable Data / Human Subjects

Q. I use personally identifiable data or data from human subjects and I obtained IRB (or similar) approval. Do I need to include proof of approval?

A. NO. It is sufficient at submission time to state that you obtained IRB approval (either in the paper or in the supplemental material). Do not include any material that would identify your institution or de-anonymize your submission in another way.

Q. I use personally identifiable data or data from human subjects, but my IRB (or similar) approval is not completed. What should I do?

A. We recommend holding off on a submission. Note also that changes in the experimental protocol mandated by the IRB can no longer be accommodated once the paper has been submitted.

Q. I use personally identifiable data or data from human subjects, yet I do not / cannot get IRB approval, e.g. as my institution does not have an IRB. What should I do?

A. Carefully describe in your paper or supplemental material how you ensure that the collection and use of data follows the ethical principles set up in the ethics guidelines. One possible path would be to obtain clearance from an independent IRB service.

Q. I have a specific application, which makes it difficult to obtain consent for data collection. What should I do?

A. Explain your situation either in your paper or in the supplemental material and ensure to include sufficient information to support that you followed the ethical principles set up in the ethics guidelines.

Q. Can my paper get rejected for an inappropriate use of personal data or data from human subjects?

A. We want to increase awareness in the CVPR community of this important issue, hence asking authors to give information on the use of such data. While we do not have a formal ethics review process, reviewers will be asked to flag any significant ethical concerns. Cases with significant ethical concerns will be referred to an ombudsperson, who will assess the situation and advise the program chairs. The program chairs reserve the right to reject papers with grave ethical issues, but expect this to occur only in exceptional circumstances.

Q. My research uses datasets that have been withdrawn by their creators, such as DukeMTMC-ReID or MS-Celeb-1M. What should I do?

A. Generally, papers should not use datasets that have been withdrawn by their creators, as doing so may involve ethical violations or even legal complications. Under some circumstances, authors may feel they need to use such datasets — for example, if fair comparison is impossible in any other way. However, authors who use such datasets should always explain the need to do so carefully and in some detail as such claims will be carefully scrutinized. Note that in many cases alternative datasets exist. The recommended course should be to not use the dataset, and (if necessary) explain that this may affect certain comparisons with prior art. It is a violation of policy for a referee or area chair to require comparison on a dataset that has been withdrawn without a detailed consultation with PCs or DEI chairs or ombuds.

Q. My research relies on broadly used public datasets of others, which have not been withdrawn, but for which it is unclear if they have been approved by an IRB. Is this allowed?

A. In the case of broadly used datasets that are still offered by their creators, for which IRB approval status is unclear, authors are encouraged to discuss the situation, e.g. why no better alternatives are available

About Potential Negative Societal Impact 

Q. Does a paper automatically get rejected if it has a potential negative impact?

A. NO, technologies often have two sides. We encourage authors to discuss this, because it is beneficial for the community to be aware of this issue.

Q. If we (authors) did not discuss potential societal impact in the paper, will the paper be automatically rejected?

A. NO, it would not be automatically rejected. However, reviewers and area chairs will be instructed to favorably consider the inclusion of a meaningful discussion. While the program chairs reserve the right to reject papers with obvious and grave negative societal impact, regardless of whether or not the issue is discussed in the paper, we expect this to occur only in exceptional situations.

 

About Limitations

Q. Does a paper automatically get rejected if it does not discuss limitations?

A. NO, it would not be automatically rejected. However, if reviewers and area chairs identified limitations that are not stated in the paper, this can be a strong negative factor that may lead to rejection.

Q. Does a paper automatically get rejected if it has certain limitations?

A. NO, it would not be automatically rejected. No technology is perfect, especially for computer vision problems that are intrinsically inverse and ill-posed. Certain assumptions need to be made, which may not be able to fully reflect all “nuisances” in the real world. The question from the reviewers’ perspective is whether the authors have a full understanding of the proposed technology, such that they know under what condition it works, and under what conditions it would not work. Real knowledge advancement comes from such understanding, which is essential for a paper to be published. 

Q. Does my paper need to have a dedicated limitations section?

A. No, a dedicated section is not required. It is perfectly acceptable to weave in the discussion of limitations throughout the paper.

 

About Code Submission

Q. Is a code submission required?

A. NO, it is completely optional. But we encourage code submissions to aid reproducibility.

Q. I cannot submit my code, e.g. because it is proprietary. What should I do?

A. We understand that code may be proprietary, preventing you from submitting it. In this case, simply state so in the submission form. If there are other reasons for not being able to submit code, you are encouraged to explain this in your submission, such as in the supplemental material. Examples include code that is not sufficiently stand-alone to include (though see below for suggestions for how to handle such cases).

Q. Does submitted code need to be anonymized?

A. CVPR is a double blind conference, so authors should make a reasonable effort to anonymize the submitted code and data. This means that author names, institution names (also in copyright / license statements) should be removed. If the paper gets accepted, we expect the authors to replace the submitted code by a non-anonymized version or link to a public GitHub repository.

Q. Are anonymous GitHub links allowed?

A. Yes. However, they have to be on a branch that will not be modified after the submission deadline. Please enter the GitHub link in a standalone text file in a submitted ZIP file.

Q. How will the submitted code be used for decision-making?

A. The submitted code will be used as additional evidence provided by the authors to add more credibility to their results. We anticipate that high-quality papers whose results are judged by our reviewers to be credible will be accepted to CVPR, even if code is not submitted. However, if something is unclear in the paper, then code, if submitted, will provide an extra chance for reviewers to verify it.

Q. If code is submitted, do you expect it to be published with the rest of the supplementary? Or, could it be withdrawn later?

A. YES, we expect submitted code to be published with the rest of the supplementary. However, if the paper gets accepted, then the authors will get a chance to update the code before it is published by adding author names, licenses, etc.

Q. Do you expect the code to be standalone? For example, what if it is part of a much bigger codebase?

A. We expect your code to be readable and helpful to reviewers in verifying the credibility of your results. It is possible to do this through code that is not standalone -- for example, with proper documentation.

Q. My code requires a dataset that is too large to include as part of the supplemental material. Can I provide an anonymous link to the dataset?

A. Similar to links to anonymous code repositories on GitHub, it is permitted to provide anonymized links to datasets that are required to execute the code. Such links must not allow to discover the authors’ identity in any way. Moreover, such anonymous data storage must not allow authors to discover who accessed the data in any way

Q. What about pseudocode instead of code? Does that count as code submission?

A. Yes, we will count detailed pseudocode as code submission as it is helpful to reviewers in validating the credibility of your results.

Q. Do you expect authors to submit data?

A. We understand that many of our authors work with highly sensitive datasets, and are not asking for private data submission. If the dataset used is publicly available, there is no need to provide it. If the dataset is private, then the authors can submit a toy or simulated dataset to illustrate how the code works.

Q. Who has access to my code? For how long?

A. Only the program chairs, the technical chairs, and the reviewers and area chairs assigned to your paper will have access to your code. We will instruct reviewers and area chairs to keep the code submissions confidential (just like the paper submissions), and delete all code submissions from their machine at the end of the review cycle. Please note that code submission is also completely voluntary.

Q. I would like to revise my code/add code during author feedback. Is this permitted?

A. Unfortunately, no. But please remember that code submission is entirely optional

 

CVPR 2022 Program Chairs can be contacted via 

programchairs-cvpr2022@googlegroups.com