Where to Submit Computer Vision Work
Published · paper counts are current, not as-published
Choosing where to send a computer-vision paper is usually discussed as a question of fit - which venue likes which kind of work. Measured against this corpus that turns out to be the wrong question, because the venues barely differ by subject. What separates them is when they run and how much they take.
Half the Calendar Is Closed in Any Given Year
4 venues in this field are indexed here, and 2 of them run in alternate years: ICCV in odd years and ECCV in even years. They do not overlap, so in any single year a researcher is choosing between 3 of them, not 4.
| Venue | Papers | Indexed editions | Years | Runs |
|---|---|---|---|---|
| CVPR | 22,520 | 14 | 2013-2026 | every year |
| ICCV | 9,145 | 7 | 2013-2025 | odd years |
| ECCV | 6,166 | 4 | 2018-2024 | even years |
| WACV | 4,435 | 7 | 2020-2026 | every year |
They Do Not Differ Much by Subject
The obvious way to choose between venues is by what they publish. That does not work here. Ranking each venue's topics and intersecting the top three leaves Multimodal Learning common to every one of them, and the lists that remain are near-permutations of one another rather than different fields of work. Whatever distinguishes these venues to a reviewer, it is not visible as a difference in subject at this level.
So the usable differences are the two the table shows: the calendar, and scale. Each venue's own page carries its full topic breakdown and year-by-year volume if you want to look closer - CVPR, ICCV, ECCV, WACV - and side-by-side comparison puts any two of them against each other.
What This Does Not Show
- Not deadlines, and not acceptance rates. The corpus records published papers. It holds no submission dates, no review timelines and no acceptance figures, which are the things a submission plan actually turns on.
- Editions indexed here, not a venue's full history. The years column is this corpus's coverage. A venue older than its first year above ran before this index reached it.
- A field is not its indexed venues. Computer-vision work also appears at the general machine-learning venues and in journals outside this set; this page compares only what is indexed under this field.
- Topic labels are assigned, not declared. They come from a classifier, and the methodology page states which model and what that costs.
The numbers here are read from the database when you load the page, so they move as the corpus does.