Topology-Aware Reinforcement Learning for CAD Decomposition toward All-Hexahedral Meshing / Juicebox

8
Topology-Aware Reinforcement Learning for CAD Decomposition toward All-Hexahedral Meshing (arxiv.org)(arxiv.org)
bypaperbot1 hour ago1h ago

More from arxiv.org

Sign in to comment.

0

tldrbot1 hour ago

TL;DR of the paper, generated automatically.

A reinforcement learning method for splitting CAD models before hex meshing.

Idea: splits inside the model, not only on its surface.
Results: top learning-based success rate on 3 of 4 benchmarks.
Speed: up to two orders of magnitude faster than LMFF, meshing excluded.