At a Glance

Battle-tested computer graphics techniques
Emphasize practicality and reproducibility over novelty and theory
Single-blind peer review with editorial oversight and shepherding
Upon acceptance, articles are immediately published online in electronic form
Articles are assigned volume, issue, and page numbers
Submissions may use any reasonable article template (e.g., ours, SIGGRAPH, EG, GPU Pro, etc.)
Authors retain copyright on their articles, but grant permission for the journal (and anyone else) to distribute them unmodified indefinitely under the CC BY-ND 3.0 license
Any source code and data provided must be under a non-restrictive Open Source license (e.g., MIT, BSD) so that it is directly usable by readers. Papers with usable source code are much more likely to be accepted.
No cost to authors or readers
Review form (updated 2016-03-11)
LaTeX template for accepted papers (updated 2016-11-28)

Format

We are now accepting single-blind submissions of formal scientific and engineering articles on computer graphics. Papers may be:

Clever hacks and tricks: previously unpublished conventional wisdom or new implementation tweaks
Novel techniques and algorithms: "battle-tested" new methods with well-analyzed characteristics, especially ones used in real production
Implementation and case studies: production notes, implementation details, discussion of practical workflow around a technique, or new and extensive analysis of previous work
Surveys: summary and discussion of a family of published techniques, similar to a textbook chapter

These may apply to any area of computer graphics, including rendering, user interfaces, modeling, animation, hardware design, geometry, topology, games, film, CAD, DCC, and visualization.

Although many JCGT papers are short ones that describe mathematically-based techniques, we also welcome "systems" papers on techniques. We recognized that such papers are necessarily larger and more complex because they encompass an entire design or implementation. Because the time to review and edit a paper is naturally proportional to its length, it may be in the authors' best interests to submit multiple short papers on the same system when the underlying elements are not tightly coupled.

All articles should be "gems" that emphasize simplicity, clarity, and utility. Their goal is to help implementers decide if the technique is appropriate for their application and then assist in that implementation.

The limitations, border cases, and artifacts of a technique should be explained in detail. For example, it is fine for a technique to work only well 5% of the time, so long as the reader understands the conditions that trigger that case and what happens in the other cases. "Battle-tested" means that you have thoroughly investigated these limits. There are many ways to accomplish this, including experiments on many and diverse scenes or platforms, careful analytic analysis of the entire input domain followed by experiments, or widespread distribution of the technique among practitioners. Production use is of course the gold standard for battle-testing: if you made a product, then it is definitely useful and something that the community wants to know about.

There is no minimum or maximum article length, however we recommend shorter articles with supplemental code, data, and video. Only directly relevant related work should be cited. Papers that include all implementation details and supplemental code are much more likely to be accepted.

Papers must be original work for which the author owns the copyright on the text and has permission to distribute the images, published source code, and published data.

Papers must be submitted for evaluation in PDF format but can be typeset in any tool and follow any reasonable article formatting including our template, or the templates of SIGGRAPH, EG, GPU Pro, etc. Note that although the format does not matter, JCGT submissions that are reworked from manuscripts originally targeting other venues should still follow our structural standards. For example, more emphasis on robustness evaluation and implementation detail than most conferences and much less introductory and related work information. We encourage taking advantage of the electronic distribution format with high-resolution color images and text that explicitly refers to the supplemental files that will be distributed alongside it.

Authors of papers that are accepted for publication in JCGT are required to convert their work to Latex and Bibtex using our template as a condition of publication. The corresponding editor, managing editor, and editor in chief will help with this process for those unfamiliar with Latex.

Process

Complete the submission form with the contact author, title, and abstract of your paper.
You will receive a confirmation e-mail with an upload link to submit a pdf of your paper and any supplemental material.
Once all files have been received, we will solicit external reviews.
The editorial board will make a determination after the reviews are received. The article will either be recommended for publication or rejected. Rejected articles may be resubmitted. Recommended articles are assigned a corresponding editor who will work with the authors to revise the manuscript. They are guaranteed to be published if the author makes the required changes.
Recommended articles may progress through several rounds of external review or editing before acceptance.
On acceptance, the board appends an article to the current issue and assigns it page numbers. They then post the article on this website. (The current issue number increments at the end of each quarter and the volume number increments at the end of each calendar year.)

JCGT board members are encouraged to submit articles of their own. Board members have no access to information about their own submissions or the identity of reviewers, and cannot affect the review process. When the editor in chief submits an article the managing editor assigns an advisory board member to act as editor in chief for that article.