AnchorFlow - GitLite

KO | EN

gitlite — search

#react #typescript #go #golang #machine-learning #rust #deep-learning #linux #docker #nodejs #kubernetes #vue #ios #openai

AnchorFlow

(★ 53)

[CVPR 2026 Oral] A training-free, mask-free framework for 3D shape editing.

파일 탐색기

최종 버전 다운로드 (.zip)

- teaser.png
- uni3d_eval.py
- - edited.png
  - src.glb
  - src.png
- - edited.png
  - src.glb
  - src.png
- - src.glb
- - hunyuandit-finetuning-flowmatching-dinol518-bf16-lr1e5-4096.yaml
  - hunyuandit-mini-overfitting-flowmatching-dinol518-bf16-lr1e4-4096.yaml
- - demo.png
- - __init__.py
  - inv_schedulers.py
  - pipelines.py
  - postprocessors.py
  - preprocessors.py
  - rembg.py
  - schedulers.py
  - surface_loaders.py
- - train_deepspeed.sh
- .gitignore
- LICENSE
- main.py
- minimal_demo.py
- minimal_demo_with_ckpt.py
- NOTICE
- README.md
- requirements.txt
- train_demo.sh
- - mesh_render.py
- anchorflow.py
.gitignore
LICENSE
README.md
requirements_cuda121.txt
requirements_cuda124.txt
torchvision_fix.py

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# AnchorFlow: Training-Free 3D Editing via Latent Anchor-Aligned Flows

<a href='https://zhenglinzhou.github.io/AnchorFlow/'><img src='https://img.shields.io/badge/Project-Page-green'></a>
<a href='https://arxiv.org/pdf/2511.22357'><img src='https://img.shields.io/badge/Technique-Report-red'></a>
<a href='https://huggingface.co/datasets/chengzgui/Eval3DEdit'><img src='https://img.shields.io/badge/Dataset-HuggingFace-yellow'></a>

This repo provides the official implementation of **AnchorFlow**, a **training-free** framework for 3D shape editing. The method performs editing directly in the 3D latent space by aligning source and target flow trajectories with latent anchors, enabling semantic-consistent, identity-preserving, and **mask-free** 3D editing across both rigid and non-rigid scenarios.

![teaser](https://raw.githubusercontent.com/ZhenglinZhou/AnchorFlow/main/assets/teaser.png)

## Installation

AnchorFlow builds upon [Hunyuan3D 2.1](https://github.com/Tencent-Hunyuan/Hunyuan3D-2.1), which serves as the base flow model. We recommend using CUDA 12.4 (as suggested in the official Hunyuan3D instructions) or CUDA 12.1.

```bash
# Create a conda environment
conda create -n anchorflow python=3.10
```

* CUDA=12.4

```bash
# Install PyTorch
pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 \
--index-url https://download.pytorch.org/whl/cu124
# Install dependencies
pip install -r requirements_cuda124.txt
pip install torch-cluster -f https://data.pyg.org/whl/torch-2.5.1+cu124.html
```

* CUDA=12.1 (also support CUDA=12.2)

```bash
# Install PyTorch
pip install torch==2.4.0 torchvision==0.19.0 torchaudio==2.4.0 \
--index-url https://download.pytorch.org/whl/cu121
pip install -r requirements_cuda121.txt
pip install torch-cluster -f https://data.pyg.org/whl/torch-2.4.0+cu121.html
```

## Usage

Run AnchorFlow with default settings. More examples can be found in `./examples`.

```bash
python3 src/anchorflow.py
```

## Prepare Editing Data

Given a `source shape` and `editing prompt`, we first construct the editing conditions, including the `source image` and `target image`. Then, we organize the inputs and perform 3D editing.

* Step1: Rendering the `source image` from the `source shape`.
  Use the **Render Multiview Images** script in [TRELLIS](https://github.com/microsoft/TRELLIS/blob/main/DATASET.md). Then select one suitable rendering as the `source image`.
* Step2: Construct the `target image` using a 2D editing model: Apply a 2D editing model (e.g., [Nano Banana](https://aistudio.google.com/models/gemini-2-5-flash-image)) to edit the `source image` according to the given `editing prompt`, producing the `target image`.
* Step3: Organize the inputs and perform 3D editing. Place all required inputs in a folder and run the 3D editing pipeline. The process requires the following files:
  * `src.glb`: the `source shape`
  * `src.png`: the `source image`
  * `edited.png`: the `target image`

You can find example setups in `./examples`.

## Evaluation

The **Eval3DEdit benchmark** is available on [Hugging Face](https://huggingface.co/datasets/chengzgui/Eval3DEdit).

```bash
hf download chengzgui/Eval3DEdit \
  --repo-type dataset \
  --local-dir Eval3DEdit
```

We provide the Uni3D/CLIP evaluation script under `eval/`.

To run the evaluator, prepare Uni3D separately and point the script to it:

```bash
git clone https://github.com/baaivision/Uni3D.git
pip install open_clip_torch trimesh pyrender Pillow numpy
```

Download the Uni3D checkpoint and EVA02 CLIP weights following the Uni3D instructions, then run:

```bash
CUDA_VISIBLE_DEVICES=0 python eval/uni3d_eval.py Eval3DEdit \
  --pred_root outputs/anchorflow \
  --pred_mesh_relpath edited.glb \
  --metadata_csv Eval3DEdit/metadata.csv \
  --uni3d_repo /path/to/Uni3D \
  --ckpt_path /path/to/uni3d/model.pt \
  --clip_weights_path /path/to/eva02_e_14_plus_laion2b_s9b_b144k.pt \
  --output_csv results/anchorflow_uni3d.csv
```

## Notes

- If you have questions or find bugs, feel free to open an issue or email the first author (zhenglinzhou@zju.edu.cn).
- If you encouter `EGL: cannot open shared object file: No such file or directory` error during rendering mesh, try to install following packages: `sudo apt-get install libegl1-mesa libgl1-mesa-glx`.

## Acknowledgements

Our repo is built on top of several several awesome projects and works, including [FlowEdit](https://github.com/fallenshock/FlowEdit), [Hunyuan3D 2.1](https://github.com/Tencent-Hunyuan/Hunyuan3D-2.1) and [EditP23](https://github.com/editp23/editp23).

## Cite

If you find AnchorFlow useful for your research and applications, please cite us using this BibTex:

```bibtex
@article{zhou2025anchorflow,
  title={AnchorFlow: Training-Free 3D Editing via Latent Anchor-Aligned Flows},
  author={Zhou, Zhenglin and Ma, Fan and Gui, Chengzhuo and Xia, Xiaobo and Fan, Hehe and Yang, Yi and Chua, Tat-Seng},
  journal={arXiv preprint arXiv:2511.22357},
  year={2025},
}
```