Ieee Transactions On Circuits And Systems For Video Technology

What It Actually Takes to Publish in TCSVT

I've been submitting to this journal for years, and the review cycle is one of the longer ones in our area. You're looking at roughly three to five months from submission to first decision if everything goes smoothly, which is rare. My most recent submission under IEEE Transactions on Circuits and Systems for Video Technology sat with reviewers for about four months before I got the first round of comments back. Some of those comments were genuinely useful, but one reviewer clearly hadn't read the methodology section carefully and asked me to justify a baseline I'd already included in my ablation study. The journal covers video coding, processing, analysis, and multimedia systems. If your work sits at the intersection of signal processing and video applications, this is the right venue. It has an impact factor in the mid-range for IEEE circuits and systems journals, and citations from industry teams at companies like Google, Apple, and Samsung are not uncommon in the accepted papers.

Understanding the Scope of Ieee Transactions On Circuits And Systems For Video Technology

The scope is narrower than the main Transactions on Circuits and Systems part I. They specifically want video-related contributions, not general signal processing papers that happen to use video as a test case. When I submitted a paper on deep learning-based error concealment, the associate editor spent more time in the initial screening checking whether the contribution was fundamentally about video rather than about neural network architecture design. The paper was desk-rejected on that basis, which was fair enough because another IEEE transaction would have been a much better fit. Video coding standards work, perceptual quality assessment, video segmentation, and multi-view processing all fall within the accepted territory. But if you're pushing the boundaries of a general deep learning framework and just happen to evaluate on videos, you're probably going to get redirected elsewhere during review even if you pass the initial screen.

The Submission Process and What Goes Wrong

Submission goes through the IEEE manuscript central portal. You upload your LaTeX source, and they generate a PDF from it rather than accepting a manually produced PDF. This means your compilation has to be clean. I lost an entire week early in my career because a missing BibTeX entry caused the source to fail compilation, and the editorial office just rejected it without notification. Now I always compile the full source on a fresh machine before hitting submit, and I double-check the generated PDF page by page. There's no separate cover letter requirement beyond what you fill in the online form, but the form asks you to suggest potential reviewers and also names ones you want excluded. Include at least three suggested reviewers who are actively publishing in the area but not recently co-authored with you. The editor will typically send the paper to two or three reviewers, and having solid suggestions actually helps because the system matches based on publication history and research interests.

Writing for This Journal Specifically

Papers in this journal tend to run long. I've seen accepted papers at 16 to 20 pages in double-column format, and the editors don't penalize length the way some venues do as long as the content justifies it. However, the experimental sections are where most papers fail. Reviewers here expect rigorous comparison against established baselines under standardized test conditions, not just qualitative results or comparisons against papers that came out the same month as yours. My paper on frame interpolation used the VFImetric benchmark, but the reviewer wanted to see results on a second dataset because he felt the first one had selection bias. I added results on the UCF-Quality and ran the evaluation script again. That took about three days of work, and it turned out the reviewer was right that the first dataset wasn't sufficient. The paper was accepted after revision, and I learned to include at least two benchmark datasets upfront instead of waiting for reviewer complaints. The math notation matters more than you'd think. IEEE has strict requirements for how equations are numbered, how variables are defined, and how vectors and matrices are typeset. If you're using LaTeX, make sure you're loading the IEEEtran class correctly and that your equation environment produces the right numbering. I've seen papers sent back for minor revision just because the equation numbers were jumping around due to a misconfigured amsthm package.

The Review Cycle and How to Respond

When you get the review comments back, the associated editor will give you a decision category: accept, minor revision, major revision, or reject. Major revision is the most common outcome for papers that are likely to be accepted eventually. Treat the comments as if they came from someone who actually understood your work, even when they clearly didn't. I've written rebuttals that were polite on the surface but essentially said "you missed this in section 3.2 of the paper" with a direct citation. That works better than being confrontational. There's no formal rebuttal phase in the initial review round, so your response is entirely in the revision itself. You need to make changes that address every substantive concern and include a point-by-point response document that the editor sees but the reviewers don't necessarily read in full. Still, write that document carefully because the editor uses it to judge whether you've addressed the issues. One thing nobody tells you about major revisions: the same reviewers often see your revised manuscript, which means if they had a fundamental objection to your approach rather than just technical concerns, a revision won't help and you're probably going to get rejected again. In that case, the honest move is to submit to a different journal rather than burn another three months on a paper with a hostile reviewer.

Accessing Published Papers

Published articles are available through the IEEE Xplore digital library. Most universities have institutional subscriptions that grant access. If you're an individual researcher without institutional access, IEEE offers individual article purchasing, but it's expensive. Preprints on arXiv are common for this venue, though the journal doesn't formally allow dual submission, so you shouldn't post a preprint until after acceptance or at least after the first round of review if your institution permits it. The journal also publishes special issues periodically, often around specific topics like high-dynamic-range video coding or low-latency video communication. These special issues can have a faster review cycle because the guest editors manage the process, but the acceptance criteria are just as strict. My co-author and I published in a special issue on efficient video coding, and the timeline from submission to acceptance was about eight weeks compared to the usual four to six months for regular issues.

When Not to Submit Here

If your work is primarily about machine learning theory with video as one of several applications, consider another venue. If you're working on real-time embedded implementation of video codecs and the paper is heavily hardware-focused, Transactions on Circuits and Systems I might be a better fit. If your contribution is purely empirical without significant algorithmic novelty, the bar for acceptance here is high and rejection is likely. The journal is selective. The acceptance rate hovers around 30 to 35 percent based on recent publication data. That's not the lowest in IEEE transactions, but it's not open either. Papers that get in usually have clean experiments, clear mathematical formulation, and a well-motivated problem statement. If your paper is missing any of those three elements, it's worth strengthening them before submission rather than wasting time on the review process.