5 ms·

I find it particularly interesting that this post steals images from my blog [1], then adjusts the saturation / brightness on them so that a reverse image searc

I find it particularly interesting that this post steals images from my blog [1], then adjusts the saturation / brightness on them so that a reverse image search thinks they are unique. Please don't do this. Folks put effort into well-illustrated explanations; if you're going to re-use content, at least put an attribution and link on there! [1] http://www.mattkeeter.com/projects/swingline/ http://www.mattkeeter.com/projects/swingline/

heydenberk 9y ago
That's incredibly frustrating and also just ... dumb? I wouldn't have thought any less of the post if it had cited a different interesting post. I would've liked it more, even!
dahart 9y ago
You're well within reasonable bounds to request that they be removed or ask for an attribution. If you created those images, they are covered by copyright, and their use of those images without permission is a copyright violation. Speaking of attribution, why doesn't that article reference Lloyd's algorithm? https://en.m.wikipedia.org/wiki/Lloyd%27s_algorithm https://en.m.wikipedia.org/wiki/Lloyd%27s_algorithm
- mkeeter 9y ago
  I'm not familiar, but from a quick review it looks like Lloyd's algorithm is a few citations down the chain. I started from [1], which is based on [2], which uses Lloyd's algorithm (check out the Related Work section in the first paper). Still, that's a good stand-alone link – I'll add it to the writeup. [1] https://www.cs.ubc.ca/labs/imager/tr/2002/secord2002b/secord.2002b.pdf https://www.cs.ubc.ca/labs/imager/tr/2002/secord2002b/secord... [2] https://www.cs.princeton.edu/courses/archive/fall00/cs597b/papers/stipple_deussen.pdf https://www.cs.princeton.edu/courses/archive/fall00/cs597b/p...
  - dahart 9y ago
    Glad to hear it! FWIW, you are using Lloyd's algorithm and the first paper is too ([1] mentions Lloyd 7 times so it's unclear why they didn't put his paper in the references section, that seems like an oversight.) Nothing wrong with using the Wikipedia article as an informal reference, but wherever you reference the other papers, considering linking to Lloyd's IEEE paper directly. Splitting the centroid computation into a two-pass accumulation is a great idea. Even though that makes for heavy shaders, I imagine it's still faster than doing a buffer read and finding the centroids in one pass via CPU? I'd also be interested in hearing how you decide on convergence. Are you stoppping after a fixed number of iterations? Or are you able to track the offsets using the GPU - not sure if that is possible since you're writing back into the VBO directly? You probably know from playing with it that Lloyd's algorithm is notorious for very long convergence times. It gets close very quickly, and then can have a super long tail with many small iterations and then sudden large changes long after it seemed like things had settled. It's super fun to watch, especially when you dynamically add points, very reminiscent of cell division -- you should make a video for your project page! You might be interested in this paper too: http://www.dgp.toronto.edu/papers/ahausner_SIGGRAPH2001.pdf http://www.dgp.toronto.edu/papers/ahausner_SIGGRAPH2001.pdf This one uses GPU Voronoi + Lloyd's algorithm to generate mosaic tilings. The neat trick here is to use a Manhattan distance instead of a Euclidean one, in order to get voronoi cells to line up in mostly rectangular grids instead of approximating a hexagonal packing.
squeakynick 9y ago
(Author here) Matt, I'm very sorry. It was a genuine mistake. I always try hard to attribute sources (as you can see from the references in the doc, and on the couple of hundred other articles on my blog). It was nothing personal, it was an oversight :) I've corrected this now on the site. Sorry again. Thanks for bringing this to my attention. Regards /\/ick
- mkeeter 9y ago
  Thanks for the quick fix; I quite enjoyed the writeup.

heydenberk 9y ago

That's incredibly frustrating and also just ... dumb? I wouldn't have thought any less of the post if it had cited a different interesting post. I would've liked it more, even!

dahart 9y ago

You're well within reasonable bounds to request that they be removed or ask for an attribution. If you created those images, they are covered by copyright, and their use of those images without permission is a copyright violation. Speaking of attribution, why doesn't that article reference Lloyd's algorithm? https://en.m.wikipedia.org/wiki/Lloyd%27s_algorithm https://en.m.wikipedia.org/wiki/Lloyd%27s_algorithm

mkeeter 9y ago
I'm not familiar, but from a quick review it looks like Lloyd's algorithm is a few citations down the chain. I started from [1], which is based on [2], which uses Lloyd's algorithm (check out the Related Work section in the first paper). Still, that's a good stand-alone link – I'll add it to the writeup. [1] https://www.cs.ubc.ca/labs/imager/tr/2002/secord2002b/secord.2002b.pdf https://www.cs.ubc.ca/labs/imager/tr/2002/secord2002b/secord... [2] https://www.cs.princeton.edu/courses/archive/fall00/cs597b/papers/stipple_deussen.pdf https://www.cs.princeton.edu/courses/archive/fall00/cs597b/p...
- dahart 9y ago
  Glad to hear it! FWIW, you are using Lloyd's algorithm and the first paper is too ([1] mentions Lloyd 7 times so it's unclear why they didn't put his paper in the references section, that seems like an oversight.) Nothing wrong with using the Wikipedia article as an informal reference, but wherever you reference the other papers, considering linking to Lloyd's IEEE paper directly. Splitting the centroid computation into a two-pass accumulation is a great idea. Even though that makes for heavy shaders, I imagine it's still faster than doing a buffer read and finding the centroids in one pass via CPU? I'd also be interested in hearing how you decide on convergence. Are you stoppping after a fixed number of iterations? Or are you able to track the offsets using the GPU - not sure if that is possible since you're writing back into the VBO directly? You probably know from playing with it that Lloyd's algorithm is notorious for very long convergence times. It gets close very quickly, and then can have a super long tail with many small iterations and then sudden large changes long after it seemed like things had settled. It's super fun to watch, especially when you dynamically add points, very reminiscent of cell division -- you should make a video for your project page! You might be interested in this paper too: http://www.dgp.toronto.edu/papers/ahausner_SIGGRAPH2001.pdf http://www.dgp.toronto.edu/papers/ahausner_SIGGRAPH2001.pdf This one uses GPU Voronoi + Lloyd's algorithm to generate mosaic tilings. The neat trick here is to use a Manhattan distance instead of a Euclidean one, in order to get voronoi cells to line up in mostly rectangular grids instead of approximating a hexagonal packing.

squeakynick 9y ago

(Author here) Matt, I'm very sorry. It was a genuine mistake. I always try hard to attribute sources (as you can see from the references in the doc, and on the couple of hundred other articles on my blog). It was nothing personal, it was an oversight :) I've corrected this now on the site. Sorry again. Thanks for bringing this to my attention. Regards /\/ick

mkeeter 9y ago
Thanks for the quick fix; I quite enjoyed the writeup.