计划执行与检查节点
executing-plans
按既定实施计划完成工作,保持过程检查和反馈。
siggraph-workflow【效率与协作】Use to plan a SIGGRAPH or SIGGRAPH Asia Technical Papers campaign backward from the deadline, covering the two annual cycles, the form-then-upload two-step submission, the rebuttal window, the Technical Papers Committee decision, conditional-accept second-stage revision, TOG camera-ready, and the in-person presentation.
独立 GitHub 快照 · 自动分类 · 以原作者文档为准
SIGGRAPH runs on two cycles a year and a compressed single-round calendar with a journal-grade
tail. Planning backward from the right deadline — and knowing which of the two conferences you are
targeting — is the highest-leverage move. Dates below are the SIGGRAPH 2026 / SIGGRAPH Asia 2026
snapshots from resources/official-source-map.md; reopen the live Important Dates page for your
cycle first.
| Cycle | Deadline (approx.) | Conference | 2026 instance |
|---|---|---|---|
| SIGGRAPH (North America) | ~January | Summer | SIGGRAPH 2026, Los Angeles, 19-23 Jul 2026; form 15 Jan, upload 23 Jan 2026 (22:00 UTC) |
| SIGGRAPH Asia | ~May | December | SIGGRAPH Asia 2026, Kuala Lumpur, 1-4 Dec 2026; Stage-2 paper 12 May, upload 13 May 2026 |
| TOG direct | rolling | (no conference talk) | ScholarOne, outside the conference cycle — less common |
A rejected paper is typically revised against the reviews and resubmitted to the next cycle, so the two deadlines form a natural half-year rhythm. Pick the cycle whose deadline your results can credibly meet — a rushed submission to the nearer deadline usually loses to a finished one at the next.
T-0 Complete submission: PDF (or MD5) + supplemental video + representative image on Linklings
T-1wk Submission form locks title/abstract/topics/ALL co-authors (earlier hard deadline)
T-2wk Freeze results; render the final results video; clear rights on every frame/figure
T-4wk Final comparisons + timings on stated hardware; ablations complete
T-8wk Draft complete in acmart (>=2.16); teaser figure locked; internal reviews
T-12wk+ Method + implementation stable; baselines reproduced; scene set assembled
The two immovable graphics-specific costs are the results video (allow real production time) and faithful baseline comparisons (allow time to build and run others' code). Neither compresses well in the last week.
Reviews released -> you see the full set
Rebuttal window (~6 weeks post-deadline): <=1000 words, text only, address factual errors
Technical Papers Committee meeting: conditional accept / reject + track (Journal vs Conference)
[If conditional] make every required change -> second-stage committee verification
Camera-ready: TOG metadata, rights/DOI, final video, permanent code archive
Presentation: at least one author presents in person at the conference
siggraph-submission.siggraph-author-response.siggraph-camera-ready.[Target cycle] SIGGRAPH 20XX (Jan) / SIGGRAPH Asia 20XX (May) / TOG-direct
[Countdown] weeks to form deadline / upload deadline
[Critical path] video production + baseline comparisons on track? yes/no
[Post-submission] rebuttal plan / conditional-accept list / camera-ready readiness
[Blockers] <ordered, dated to the UTC cutoff>