A paper submitted to a conference can be accepted either as an oral talk, a poster talk, or a short demo/industry track, with variations (small talks, long talks, plenary talks, e-posters, etc.). In some fields, the poster is considered less important, and will not be published as a paper in proceedings. In others, they have the same importance.
Conferences have two main purposes (I forgot about vacations in paradise places):
- present your paper, either talk or poster,
- attend other talks/posters, meet people.
To me, "submitting a paper to a conference implies a moral obligation for one of the authors to present the talk or the poster, if the paper is accepted". If one does not attend, the submission should be withdrawed, for different potential reasons:
- you (and your co-authors) do not need, or cannot afford a poster presentation,
- the conference program is not good enough for your purposes,
- you should not fool attendees with papers that will not be presented.
In the conferences I do summit to, both oral and poster are considered even. I now do:
- prefer posters when I present, as I have more time (1h-2h vs 15 min-20 min) to present, to talk with attendees, adapt to their background,
- prefer attend to posters than to orals, except when I spot a specific point for a given talk.
Plus, I do bring along with me small copies of the poster, copies of papers that are related to the work at hand, that I can offer to attendees. Reminder of the root: a "publication" means: make a research "public". Posters are a good vector for that, and bring you more feedback (very few questions asked in traditional oral sessions). Sometimes, somebody who cannot attend your allotted slot can propose a specific meeting. I should admit I even went to a person presenting a poster, to ask him about her/his previous work.
In those conferences, poster chairs spot "no shows": poster panels without posters, posters without presenters, posters with a presenter who is not a co-author.
Example: No-show papers are defined as papers submitted by
authors who subsequently did not present the paper in-person (no
videos, no remote cast) at the technical meeting. Presentations by
proxies are not allowed, unless explicitly approved before the
conference by the technical co-chairs. No-show papers that were not
withdrawn and were published in the Proceedings must be identified as
"No-Show" in the files submitted to IEEE for further publication (IEEE
Xplore). No-shows will not be available on IEEE Xplore or other public
access IEEE forums. IEEE will maintain an archive of no-shows.
Authors of "no show" are sometimes blacklisted, and can be banned of publications in the scientific organization places (future conferences, even journals) for a couple of years.
So it will be perceived poorly if you do not warn the organizers and withdraw the paper, is unethical to other attendees, and involves consequences for you and your co-authors.