FM_nodes

★ 8

人脸修复视频超分图像风格融合ComfyUI节点

FM_nodes是一组为ComfyUI提供的节点集合，包含WFEN（人脸超分/修复）、RealViFormer（真实视频超分）、ProPIH（渐进画风融合）等，便于在工作流中直接调用模型与预置流程。

💡 在ComfyUI中快速应用人脸修复、视频超分与风格融合

🍴 3 Forks💻 Python🔄 2025-03-27

🔗 GitHub 原文

📦

网盘下载

复制链接后前往夸克网盘下载

https://pan.quark.cn/s/e58c8376a81b

📦 requirements.txt

torch
einops

📄 README

FM_nodes

A collection of ComfyUI nodes.

Click name to jump to workflow

WFEN Face Restore. Paper: Efficient Face Super-Resolution via Wavelet-based Feature Enhancement Network

RealViformer – Paper: Investigating Attention for Real-World Video Super-Resolution

ProPIH. Paper: Progressive Painterly Image Harmonization from Low-level Styles to High-level Styles

CoLIE. Paper: Fast Context-Based Low-Light Image Enhancement via Neural Implicit Representations

VFIMamba. Paper: Video Frame Interpolation with State Space Models

ConvIR. Paper: Revitalizing Convolutional Network for Image Restoration

StabStitch. Paper: Eliminating Warping Shakes for Unsupervised Online Video Stitching

Workflows

WFEN

Download the model here and place it in models/wfen/WFEN.pth.

workflow_wfen_facecrop.json

RealViformer

Download the model here and place it in models/realviformer/weights.pth.

workflow_realviformer.json

(Not a workflow-embedded image)

https://github.com/user-attachments/assets/e89003c0-7be5-4263-b281-fd609807cea1

RealViFormer upscale example

ProPIH

Download the vgg_normalised.pth model in the Installation section and latest_net_G.pth in the Train/Test section

models/propih/vgg_normalised.pth
models/propih/latest_net_G.pth

workflow_propih.json

CoLIE

No model needed to be downloaded. Lower loss_mean seems to result in brighter images. Node works with image and batched/video.

workflow_colie_lowlight.json

VFIMamba

Download the models from the huggingface page

models/vfimamba/VFIMamba_S.pkl
models/vfimamba/VFIMamba.pkl

You will need to install mamba-ssm, which does not have a prebuilt Windows binary. You will need:

triton. Prebuilt for Python 3.10 and 3.11 can be found here: https://github.com/triton-lang/triton/issues/2881 – https://huggingface.co/madbuda/triton-windows-builds/tree/main

causal-conv1d. Follow this post: https://github.com/NVlabs/MambaVision/issues/14#issuecomment-2232581078

mamba-ssm. Follow this tutorial: https://blog.csdn.net/yyywxk/article/details/140420538. Fork that followed all the steps: https://github.com/FuouM/mamba-windows-build

I’ve built mamba-ssm for Python 3.11, torch 2.3.0+cu121, which can be obtained here: https://huggingface.co/FuouM/mamba-ssm-windows-builds/tree/main

To install, pip install [].whl

workflow_vfi_mamba.json

(Not a workflow-embedded image)

https://github.com/user-attachments/assets/be263cc3-a104-4262-899b-242e9802719e

VFIMamba Example (top: Original, bottom: 5X, 20FPS)

ConvIR

Download models in the Pretrained models – gdrive section

workflow_convir.json

models\convir
│ deraining.pkl
│
├─defocus
│   dpdd-base.pkl
│   dpdd-large.pkl
│   dpdd-small.pkl
│
├─dehaze
│   densehaze-base.pkl
│   densehaze-small.pkl
│   gta5-base.pkl
│   gta5-small.pkl
│   haze4k-base.pkl
│   haze4k-large.pkl
│   haze4k-small.pkl
│   ihaze-base.pkl
│   ihaze-small.pkl
│   its-base.pkl
│   its-small.pkl
│   nhhaze-base.pkl
│   nhhaze-small.pkl
│   nhr-base.pkl
│   nhr-small.pkl
│   ohaze-base.pkl
│   ohaze-small.pkl
│   ots-base.pkl
│   ots-small.pkl
│
├─desnow
│   csd-base.pkl
│   csd-small.pkl
│   snow100k-base.pkl
│   snow100k-small.pkl
│   srrs-base.pkl
│   srrs-small.pkl
│
└─modeblur
    convir_gopro.pkl
    convir_rsblur.pkl

StabStitch

Download all 3 models in the Code – Pre-trained model section.

models/stabstitch/temporal_warp.pth
models/stabstitch/spatial_warp.pth
models/stabstitch/smooth_warp.pth

Use interpolate_mode = NORMAL or do_linear_blend = True to eliminate dark borders. Inputs will be resized to 360×480. Recommends using StabStitch Crop Resize.

| StabStitch | StabStitch Stabilize |

|-|-|

| stabstitch_stitch.json (Example videos in examples\stabstitch) | stabstich_stabilize.json |

| | |

(Not workflow-embedded images)

Credits

@misc{chobola2024fast,
      title={Fast Context-Based Low-Light Image Enhancement via Neural Implicit Representations}, 
      author={Tomáš Chobola and Yu Liu and Hanyi Zhang and Julia A. Schnabel and Tingying Peng},
      year={2024},
      eprint={2407.12511},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2407.12511}, 
}

@misc{zhang2024vfimambavideoframeinterpolation,
      title={VFIMamba: Video Frame Interpolation with State Space Models}, 
      author={Guozhen Zhang and Chunxu Liu and Yutao Cui and Xiaotong Zhao and Kai Ma and Limin Wang},
      year={2024},
      eprint={2407.02315},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2407.02315}, 
}

@article{cui2024revitalizing,
  title={Revitalizing Convolutional Network for Image Restoration},
  author={Cui, Yuning and Ren, Wenqi and Cao, Xiaochun and Knoll, Alois},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2024},
  publisher={IEEE}
}

@inproceedings{cui2023irnext,
  title={IRNeXt: Rethinking Convolutional Network Design for Image Restoration},
  author={Cui, Yuning and Ren, Wenqi and Yang, Sining and Cao, Xiaochun and Knoll, Alois},
  booktitle={International Conference on Machine Learning},
  pages={6545--6564},
  year={2023},
  organization={PMLR}
}

@article{nie2024eliminating,
  title={Eliminating Warping Shakes for Unsupervised Online Video Stitching},
  author={Nie, Lang and Lin, Chunyu and Liao, Kang and Zhang, Yun and Liu, Shuaicheng and Zhao, Yao},
  journal={arXiv preprint arXiv:2403.06378},
  year={2024}
}