Machine Learning

1 readers

1 users here now

Community Rules:

Be nice. No offensive behavior, insults or attacks: we encourage a diverse community in which members feel safe and have a voice.
Make your post clear and comprehensive: posts that lack insight or effort will be removed. (ex: questions which are easily googled)
Beginner or career related questions go elsewhere. This community is focused in discussion of research and new projects that advance the state-of-the-art.
Limit self-promotion. Comments and posts should be first and foremost about topics of interest to ML observers and practitioners. Limited self-promotion is tolerated, but the sub is not here as merely a source for free advertisement. Such posts will be removed at the discretion of the mods.

founded 1 year ago

MODERATORS

communick@academy.garden

[D] [P] Images to triangles (alien.top)

submitted 11 months ago by Existing-Pie-3723@alien.top to c/machinelearning@academy.garden

4 comments fedilink hide all child comments

Images to triangles

Say you have an image, which is the input ... You now need to output a set of n triangles that will most closely replicate it, where n is fixed. How will you go about this problem today

Genetic Algos have kind of tackled this problem, but takes forever or converges easily to some picture ... I tried running some preliminary things here on Neat playground here .... https://jerryjohnthomas.github.io/30pieces/
Each triangle is something takes space and color ... So try to group pixels of same color as triangle ... Like a knn thing

I think we could maybe do a much better job without cnns, adding a simplifying assumption say the image indeed is made of n triangles and exact replication is possible. Is there anything else you would try.

top 4 comments

sorted by: hot top controversial new old

[–] 0xcedbeef@alien.top 1 points 11 months ago

Make a voronoi diagram
Triangulate the voronoi cells.
Make the triangles nice (do delaunay flips). I think it's ok if you break the voronoi cells
Color

No ML needed.

The way you sample at (1) will affect your result, you probably don't want to sample where there's edges in the image

[–] Real_Revenue_4741@alien.top 1 points 11 months ago

Honestly, RL sounds like a decent approach here.

[–] bablador@alien.top 1 points 11 months ago

Simulated annealing

[–] currentscurrents@alien.top 1 points 11 months ago

I have done something similar using gradient descent and this differentiable vector graphics library. It converges much faster than genetic algorithms.

A good initialization, like the other commenter's voronoi idea, would speed it up considerably.