---
language:
- en
license: mit
size_categories:
- n<1K
pretty_name: Edit3D-Bench
tags:
- 3D
- Editing
task_categories:
- image-to-3d
---

This repository contains Edit3D-Bench, a 3D editing benchmark proposed in the paper [Feedforward 3D Editing via Text-Steerable Image-to-3D](https://huggingface.co/papers/2512.13678).

Project page: https://glab-caltech.github.io/steer3d/
Code: https://github.com/ziqi-ma/Steer3D

The `metadata/` directory stores metadata information (the source and target object guid, and editing text) for texture, removal, and addition - each in a separate `.csv` file.
The `data/` directory contains source images (of unedited object), glbs of both source and target objects, and [TRELLIS](trellis3d.github.io) latents of both source and target objects, each indexed with the object guid.

### Sample Usage

This dataset is designed to benchmark 3D editing capabilities. To use it for evaluation with the associated [Steer3D codebase](https://github.com/ziqi-ma/Steer3D), follow these steps:

1.  **Clone the dataset:**
    ```bash
    git lfs install
    git clone https://huggingface.co/datasets/ziqima/Edit3D-Bench
    ```

2.  **Set up the Steer3D environment:**
    The Steer3D model requires a specific environment. Refer to the [Steer3D GitHub repository](https://github.com/ziqi-ma/Steer3D) for the latest setup instructions. A typical setup involves:
    ```bash
    conda env create -f environment.yml
    conda activate steer3d
    ```
    Note that libraries `kaolin`, `nvdiffrast`, `diffoctreerast`, `mip-splatting`, and `vox2seq` might need manual installation. Please refer to [this setup script from TRELLIS](https://github.com/microsoft/TRELLIS/blob/main/setup.sh) for installation of these dependencies.

3.  **Evaluate on the Benchmark (Texture Example):**
    Once the dataset is cloned and the Steer3D environment is active, you can run evaluation scripts. First, ensure `PYTHONPATH` is set to the path of your Steer3D clone. Then, update the `val_dataset` path in `configs/stage3_controlnet.json` within the Steer3D repository to `[path-to-Edit3D-Bench-clone]/metadata/texture.csv`.

    ```bash
    python inference/inference_texture.py \
            --stage1_checkpoint [path-to-checkpoints]/stage1/base.pt \
            --stage1_config configs/stage1_controlnet.json \
            --stage2_controlnet_checkpoint [path-to-checkpoints]/stage2/controlnet.pt \
            --stage2_base_checkpoint [path-to-checkpoints]/stage2/base.pt \
            --stage2_config configs/stage2_controlnet.json \
            --output_dir visualizations/output \
            --num_examples 150 \
            --num_seeds 3 \
            --split val
    ```

### Citation

If you find our work helpful, please cite using the following BibTeX entry:

```bibtex
@misc{ma2025feedforward3deditingtextsteerable,
      title={Feedforward 3D Editing via Text-Steerable Image-to-3D}, 
      author={Ziqi Ma and Hongqiao Chen and Yisong Yue and Georgia Gkioxari},
      year={2025},
      eprint={2512.13678},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2512.13678}, 
}
```