TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Derivatives and Logarithms of 3D Transforms

105 pointsby nosferalatu1232 months ago

6 comments

ndriscoll2 months ago
In addition to the differential equation, you can also tweak the definition of the exponential function as a limit used with e.g. compound interest:<p><pre><code> exp(tA) = lim n-&gt;infinity (I+tA&#x2F;n)^n = lim n-&gt;infinity (I+tA&#x2F;n)...(I+tA&#x2F;n) (n times) </code></pre> So you can interpret A (or log T) as a direction to move from the identity, and exp does infinite iterated compositions of an infinitesimal shift away from the identity in that direction.
MITSardine2 months ago
Cool article. Regarding section &quot;The exponential map and logarithm map&quot;, if you&#x27;re interested in computing the matrix exponential, there is the classic: C Moler, C Van Loan, &quot;Nineteen dubious ways to compute the exponential of a matrix, twenty-five years later&quot;. Also, using the series expansion is not necessarily unrobust as long as you don&#x27;t stop at a fixed number of iterations but instead go on as long as terms have a norm greater than some tolerance. Scaling and squaring can be used to remain always in a given range of norms (less than 1, say).<p>Regarding Pitfall #3, the interpolation scheme exp(tlog(A) + (1-t)log(B)) is shortest path in a sense, just not with the usual matrix norms. See V Arsigny et al., &quot;Log‐Euclidean metrics for fast and simple calculus on diffusion tensors&quot;. I can&#x27;t help but find it more elegant than exp(log(BA^{-1})t)A which could just as well have been exp(log(A^{-1}B)t)A, or even Aexp(log(A^{-1}B)t), right? It also fixes the &quot;no more than two transforms&quot;, as you can put any convex combination in exp(sum_i x_i log(A_i)).
lenkite2 months ago
Just amazed that applets are still being used and are still working. Many folks here are younger than Java Applet technology.
评论 #43452853 未加载
评论 #43451922 未加载
moktonar2 months ago
What if you spherically interpolate the basis vectors and linearly interpolate the translation vector?
xeonmc2 months ago
essentially you decompose the transformation into (axial translation) + (screw rotation) + (oriented orthogonal stretch) and each of them are just straightforward interpolation: axial is linear, screw is angular, and stretch is exponential.
评论 #43451011 未加载
imtringued2 months ago
See also:<p>Screw Theory: <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Screw_theory" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Screw_theory</a>