Skip navigation
1-layer, 8-bits parameter neural net doing pattern completion with sparse distributed activations…
0:00 / 1:12

1-layer, 8-bits parameter neural net doing pattern completion with sparse distributed activations…

Jeacom

@Jeacom2563.6K followers

Visit on X
206 likesDownloadOriginal post

22,379 views #b3d #blender #retopology

1-layer, 8-bits parameter neural net doing pattern completion with sparse distributed activations and grid-cell-like vector fields can learn to retopo, maybe. #b3d #blender #retopology

7 Comments

Comments from the original post on X (Twitter)

  • Jeacom's profile picture

    @Jeacom256

    Still having trouble with the brush coordinates thing.

    2 likes

  • Daniel Kreuter's profile picture

    @dani_kreuter

    I love this idea! I’m looking forward to seeing what will become this concept 😊

    3 likes

  • L̟̹̩̾̐apisSea🔶's profile picture

    @TheLapisSea

    In my experience retopology is not really an issue of how to place the next quad in a grid, it's more an issue of how to achieve the wanted edge flow and density in certain area while minimizing poles or triangles and maximizing the preserved detail of the base shape.

    2 likes

  • Jeacom's profile picture

    @Jeacom256

    well, this is gonna be a quad brush and not a auto-retopo button so unfortunately, solving poles is a task for the user. but I guess automatically placing and connecting dozens of quads in sensible locations with a single brush stroke is going to release a bit of the burden.

    7 likes

  • AFX's profile picture

    @AFX_LAB

    This looks awesome 👏

    1 like

  • ADecadeAtMost's profile picture

    @JanssenJiels

    How’s Blender sending and receiving data to the model and whete is the model running?

  • Jeacom's profile picture

    @Jeacom256

    The model is running on CPU locally inside blender's thread. its all mostly custom un-optmized python code and some Cython for vector crunching.

    3 likes