[Soc-2016-dev] Weekly Report #9, UV-Tools

[Phil Gosch]

Report #9 for UV-Tools:

What I did this week:

* Spend quite some time with finishing the implementation of Simulated Annealing, the metaheuristic we use to iteratively find the optimal packing solution. This also included re-reading papers and refactoring some of the code to follow the implementations used in those papers. I also implemented collision-free-region computation according to [1]. This turned out to be slower and unnecessary complicated for now, so I reverted back to the random position sampling I was using before. I kept the implementation around however, it will come in handy once I get to implement concave boundary support for UV islands/charts.

* Started implementing support for margin/padding.

* Discovered and fixed bugs for other tools, for example "Select overlapping UVs".

What I plan on doing next week:

* While talking to my mentor today I realized a few things I need to adjust/refactor in the current SA implementation to make it work as intended. I hope to have these finished soon (along with margin support), so I can start providing testbuilds and make sure I have a stable operator users can work with before starting to add support for concave bounds and holes.

Questions:

All the questions from last reports are still unanswered.

[1] 
http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1678-58782008000300005

-- 
pixel-pusher at saphirestudio.at