---
title: ComfyUI-pixel-perfect-depth
canonical_url: "https://www.modelscope.cn/models/ApacheOne/ComfyUI-pixel-perfect-depth"
md_url: "https://www.modelscope.cn/models/ApacheOne/ComfyUI-pixel-perfect-depth.md"
repository: ApacheOne/ComfyUI-pixel-perfect-depth
last_updated: 2026-02-20
license: apache-2.0
pipeline_tag: image-to-3D
tasks:
  - image-to-3D
model_type:
  - custom
parameters: 958.7M
tensor_type:
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 47
stars: 0
tags:
  - art
---

# ComfyUI-pixel-perfect-depth

> ComfyUI-pixel-perfect-depth - ApacheOne 在 ModelScope 开源的模型。Working on tonight the RT versions Model Downloads for ComfyUI custom node Here Node usage on github Orgin models can be found linked here at orgin github in google drives or some at these…

ApacheOne/ComfyUI-pixel-perfect-depth 是 ModelScope 魔搭社区上的 958.7M 参数image-to-3D模型，采用 apache-2.0 许可。

- **Repository**: ApacheOne/ComfyUI-pixel-perfect-depth
- **License**: apache-2.0
- **Tasks**: image-to-3D
- **Parameters**: 958.7M
- **Tags**: art
- **Downloads**: 47
- **Stars**: 0
- **Last updated**: 2026-02-20

Source: https://www.modelscope.cn/models/ApacheOne/ComfyUI-pixel-perfect-depth

---

![PPD sample output](https://raw.githubusercontent.com/Apache0ne/ComfyUI-pixel-perfect-depth/main/examples/ImageToStl.com_PPD_00020_.ply.gif)

- Working on tonight the RT versions 
- Model Downloads for ComfyUI custom node [Here](https://github.com/Apache0ne/ComfyUI-pixel-perfect-depth)
  Node usage on github
- Orgin models can be found linked [here](https://github.com/gangweix/pixel-perfect-depth) at orgin github in google drives or some at these Huggingfaces listed below

- https://huggingface.co/gangweix/Pixel-Perfect-Depth
- https://huggingface.co/depth-anything/Depth-Anything-V2-Large
- https://huggingface.co/Ruicheng/moge-2-vitl-normal
- https://huggingface.co/yyfz233/Pi3

## Acknowledgement

We are grateful to the [Depth Anything V2](https://github.com/DepthAnything/Depth-Anything-V2), [MoGe](https://github.com/microsoft/MoGe) and [DiT](https://github.com/facebookresearch/DiT) teams for their code and model release. We would also like to sincerely thank the NeurIPS reviewers for their appreciation of this work (ratings: 5, 5, 5, 5).

## Citation

If you find this project useful, please consider citing:

```bibtex
@article{xu2025pixel,
  title={Pixel-perfect depth with semantics-prompted diffusion transformers},
  author={Xu, Gangwei and Lin, Haotong and Luo, Hongcheng and Wang, Xianqi and Yao, Jingfeng and Zhu, Lianghui and Pu, Yuechuan and Chi, Cheng and Sun, Haiyang and Wang, Bing and others},
  journal={arXiv preprint arXiv:2510.07316},
  year={2025}
}