brycewang-stanford WACV Workflow
Use this to run a WACV campaign end to end. The defining feature is that WACV has two rounds with a revision path between them, so the calendar is a branching tree, not a single deadline. Dates below are the reported WACV 2027 timetable read on 2026-07-09 (via search rendering — reconfirm on the live Dates page); the structure is stable, the exact dates are not.
The branching calendar
Track + round decision ──► Round 1 paper deadline (reported Jun 26, 2026)
│
├─ R1 reviews (reported due Jul 24) ─► optional 1-page rebuttal
│
└─ R1 decision (reported Aug 7)
├─ Accept ──────────────► camera-ready
├─ Reject ──────────────► re-target / rebuild
└─ Revise & Resubmit ─► revise + change summary
│
Round 2 deadline (reported Aug 28) ◄─ new R2 papers also enter here
│
R2 reviews (reported due Sep 25) ─ (no rebuttal)
│
R2 decision (reported Oct 9) ─► Accept / Reject
│
registration (reported Nov 2) ─► winter conference (Jan 2027, 待核实)
Phase-by-phase
| Phase | Do this | Skill |
|---|---|---|
| Scope | Decide WACV vs sibling, Applications vs Algorithms, and R1 vs R2 | wacv-topic-selection |
| Draft | Write the first page for the chosen track and the two-round reviewer | wacv-writing-style |
| Evidence | Build matched-baseline, constraint-aware experiments with uncertainty | wacv-experiments, wacv-reproducibility |
| Submit R1 | Audit anonymity, profiles, page cap, track field | wacv-submission, wacv-supplementary |
| R1 reviews | Draft the optional one-page rebuttal, AC-first | wacv-author-response, wacv-review-process |
| R&R lap | Execute the revision as a spec; write the change summary; sync the artifact | wacv-author-response, wacv-artifact-evaluation |
| Camera-ready | De-anonymize, finalize for CVF open access + IEEE Xplore | wacv-camera-ready |
| Present | Prepare the winter conference talk/poster | wacv-camera-ready |
Branch points a single-deadline plan misses
- Round choice is a scheduling decision, not just a deadline. A borderline paper belongs in Round 1 (rebuttal + Revise-and-Resubmit net); a polished one can go to Round 2. Deciding this late forfeits the safety net.
- The Revise-and-Resubmit lap is the main event, not a detour. Most WACV papers pass through it; budget real engineering time for the revision between the R1 decision and the R2 deadline, because Round 2 has no rebuttal to compensate for a rushed revision.
- Two OpenReview groups. Round 1 and Round 2 are distinct groups; do not assume a single submission carries over automatically — follow the current instructions for resubmitting a Revise-and-Resubmit paper.
A worked mini-plan (borderline applications paper)
T-3 wk: freeze track = Applications; target Round 1 for the R&R net
T-2 wk: matched-budget baselines re-run; uncertainty over seeds/sessions locked
T-1 wk: anonymity + profile + page-cap audit (wacv-submission)
R1 deadline: submit
+reviews: one-page rebuttal fixes the top factual objection
R1 = Revise & Resubmit: turn the three reviews into a change list; execute; write summary
R2 deadline: resubmit revised paper + change summary
R2 = Accept: camera-ready; register by the deadline; prep winter poster
Reverify each cycle
- Both round deadlines, the review/decision dates, and the registration deadline.
- Whether the two-round model and the three R1 outcomes still hold.
- The camera-ready deadline and the winter conference dates/location (2027 dates 待核实).
Output format
[Current phase] scope / draft / R1 submit / R1 reviews / R&R / R2 / camera-ready / present
[Round + track] R1|R2 · Applications|Algorithms
[Next hard date] <deadline> — <what it gates>
[Branch risk] <the decision that, if made late, forfeits the R&R net>
[Next skill] <which WACV skill to open now>