Skip to content

Mesh collision loss, Deep learnning, Neural network, Pytorch, Python

Notifications You must be signed in to change notification settings

yeyuxmf/mesh_collision_loss

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 

Repository files navigation

mesh_collision_loss

  1. This is only a simple application, not an accurate calculation. You can make modifications according to your own needs.

Implementation principle:

  1. Here, collisions between different meshes are transformed into collisions between a group of point clouds and meshes.
  2. Secondly, the distance between points and faces is converted into the distance between the point and the center point of the face.
  3. This operation is simple, fast, and experimental results show effectiveness.

Visualization Tool Installation Package

链接:https://pan.baidu.com/s/1SMPms9_GoujxcmbTCV0K3Q (Geomagic Wrap 2013 (64 bit)) 提取码:fiid
This is a 3D visualization tool that supports visualization of stl files and point cloud txt file formats.

Experimental result

Experimental loss

Reference

  1. https://docs.torchdrivesim.org/en/latest/_modules/torchdrivesim/infractions.html
  2. https://github.com/facebookresearch/pytorch3d

Environment

1.python 3.7.0

2.pytorch 11.3.1

3.pytorch3D 0.7.4

4.vedo 2023.4.4

Train

python ./mesh_collision_loss.py

License and Citation

1.Without permission, the design concept of this model shall not be used for commercial purposes, profit seeking, etc.

2.If you refer to the design concept of this model for theoretical research, please also add a reference.

About

Mesh collision loss, Deep learnning, Neural network, Pytorch, Python

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages