ComfyUI_OmniPart

ComfyUI_OmniPart
★ 9

3D生成部件感知语义解耦ComfyUI节点
在ComfyUI中使用OmniPart进行部件感知的3D生成,实现语义解耦与结构一致性,支持分步标注与手动合并,便于生成可控的零部件级三维结果。
💡 生成具有部件级控制的三维模型,分两步运行并手动合并结果
🍴 1 Forks💻 Python🔄 2025-10-15
📦
网盘下载
复制链接后前往夸克网盘下载
https://pan.quark.cn/s/c1eafc754fbb
📦 requirements.txt
--extra-index-url
https://download.pytorch.org/whl/cu121
torch==2.4.0
torchvision==0.19.0
pillow==10.4.0
imageio==2.36.1
imageio-ffmpeg==0.5.1
tqdm==4.67.1
easydict==1.13
opencv-python-headless==4.10.0.84
scipy==1.14.1
rembg==2.0.60
onnxruntime==1.20.1
trimesh==4.5.3
xatlas==0.0.9
pyvista==0.44.2
pymeshfix==0.17.0
igraph==0.11.8
git+https://github.com/EasternJournalist/utils3d.git@9a4eb15e4021b67b12c460c7057d642626897ec8
xformers==0.0.27.post2
spconv-cu120==2.3.6
transformers==4.50.3
pydantic==2.10.6
diffusers==0.32.0
lightning==2.2
mesh2sdf
loguru
tetgen==0.6.3
pymeshfix
igraph
omegaconf
pycocotools
kornia
timm
h5py
boto3
git+https://github.com/facebookresearch/segment-anything.git
git+https://github.com/facebookresearch/detectron2.git
--find-links
https://data.pyg.org/whl/torch-2.8.0+cu128.html
torch-scatter
https://github.com/Dao-AILab/flash-attention/releases/download/v2.7.0.post2/flash_attn-2.7.0.post2+cu12torch2.4cxx11abiFALSE-cp310-cp310-linux_x86_64.whl
https://huggingface.co/spaces/JeffreyXiang/TRELLIS/resolve/main/wheels/diff_gaussian_rasterization-0.0.0-cp310-cp310-linux_x86_64.whl?download=true
https://huggingface.co/spaces/JeffreyXiang/TRELLIS/resolve/main/wheels/nvdiffrast-0.3.3-cp310-cp310-linux_x86_64.whl?download=true
📄 README

ComfyUI_OmniPart

OmniPart: Part-Aware 3D Generation with Semantic Decoupling and Structural Cohesion,this node ,you can use it in comfyUI

Upadte

  • test cu128 torch2.8.0 Vram 12G
  • 因为要手动标记合并,所以要分两步运行,不知道就看我B站视频吧。
  • 1.Installation


    In the ./ComfyUI/custom_nodes directory, run the following:

    git clone https://github.com/smthemex/ComfyUI_OmniPart

    2.requirements


  • 比较难装,当然你有安装trellis,基本上没什么难度,除了要再装一个seg和detectron2
  • pip install -r requirements.txt

    3.checkpoints


  • 3.1 OmniPart repo 微调的trellis模型?可能也没微调
  • 3.2 omnipart/OmniPart_modules all checkpoints 目录下三个模型
  • 3.3 dinov2 only model.safetensors 只下model.safetensors and links dinov2_vitl14_reg4_pretrain.pth;
  • 3.4 RMBG-2.0 only model.safetensors 只下model.safetensors;
  • ├── ComfyUI/models/OmniPart
    |     ├── bbox_gen.ckpt
    |     ├── model.safetensors #RMBG-2.0
    |     ├── partfield_encoder.ckpt
    |     ├── sam_vit_h_4b8939.pth
    ├── ComfyUI/models/dinov2
    |        ├──model.safetensors #facebook/dinov2-with-registers-large
    |        ├──dinov2_vitl14_reg4_pretrain.pth
    ├── any-path/omnipart/OmniPart  # https://huggingface.co/omnipart/OmniPart/tree/main 全下,保持目录一致,几G而已  

    Example

    Acknowledgements

    TRELLIS

    PartField

    FlexiCubes

    Citation