Wikipedia:Articles for deletion/Generalized quaternion interpolation: Difference between revisions
Content deleted Content added
Line 10:
<div class="xfd_relist" style="border-top: 1px solid #AAA; border-bottom: 1px solid #AAA; padding: 0px 25px;"><span style="color: #FF4F00;">'''{{resize|91%|[[Wikipedia:Deletion process#Relisting discussions|Relisted]] to generate a more thorough discussion so a clearer consensus may be reached.}}'''</span><br />
<small>Please add new comments below this notice. Thanks, <span class="smallcaps" style="font-variant:small-caps;">[[User:Northamerica1000|North America]]<sup>[[User talk:Northamerica1000|<font size="-2">1000</font>]]</sup></span> 00:34, 21 April 2015 (UTC)</small><!-- from Template:Relist -->[[Category:Relisted AfD debates|Generalized quaternion interpolation]]</div>
*'''Delete''' I found the (single) reference online: [https://hal.archives-ouvertes.fr/inria-00073318/document]. The whole thing looks borderline [[WP:NOTESSAY]] and complete gibberish. I first tried to understand what this article is about without the reference with my smattering of elemental mathematics, and frankly the most basic things are not correctly explained. Some basic and probably incorrect summary of the reference follows with as little jargon as possible:
:::The main problem is to ''measure a 3d rotation''. We have multiple measurements with uncertainty attached to each of them and we want to take a guess of what the real value is. Notice this "value" is a three-parameter thing, for instance axis of rotation (2 degrees of freedom) and angle (1 DoF), so it can be represented by a unit vector of the quaternion space (if you forget about compositions of rotations, that's equivalent to a [[3-sphere]]).
:::The "naive" way to look at the problem is to use some weighted average of the measurements (it is already not that easy if measurements have inconsistent error bars). But the thing is, that average is not easy to define, for instance the average of a set of unit vectors is not a unit vector, so you cannot find a straightforward geometrical definition for "average" here, because there are additional constraints on our objects (they must fit on a sphere). The article then proceeds to describe an algorithm that supposedly finds a good solution to the problem for a reason I do not quite see.
:Even if the article was rewritten into a clear, concise and correct summary of that reference, I am still not seeing how this could possibly be considered notable. [[User:Tigraan|Tigraan]] ([[User talk:Tigraan|talk]]) 15:35, 22 April 2015 (UTC)
| |||