Shap E

🚀 Shap-E

Shap-E introduces a diffusion process that can generate a 3D image from a text prompt, offering a new way for text-to-3D generation.

🚀 Quick Start

Shap-E is a model that can generate 3D images from text prompts. It was introduced in Shap-E: Generating Conditional 3D Implicit Functions by Heewoo Jun and Alex Nichol from OpenAI. The original repository of Shap-E can be found here.

✨ Features

• Text-to-3D Generation: Generate 3D images from text prompts.
• Multiple Representations: Directly generate the parameters of implicit functions that can be rendered as both textured meshes and neural radiance fields.

📦 Installation

First make sure you have installed all the dependencies:

pip install transformers accelerate -q
pip install git+https://github.com/huggingface/diffusers@@shap-ee

💻 Usage Examples

Basic Usage

import torch
from diffusers import ShapEPipeline
from diffusers.utils import export_to_gif


ckpt_id = "openai/shap-e"
pipe = ShapEPipeline.from_pretrained(repo).to("cuda")


guidance_scale = 15.0
prompt = "a shark"
images = pipe(
    prompt,
    guidance_scale=guidance_scale,
    num_inference_steps=64,
    size=256,
).images

gif_path = export_to_gif(images, "shark_3d.gif")

📚 Documentation

Introduction

The abstract of the Shap-E paper:

We present Shap-E, a conditional generative model for 3D assets. Unlike recent work on 3D generative models which produce a single output representation, Shap-E directly generates the parameters of implicit functions that can be rendered as both textured meshes and neural radiance fields. We train Shap-E in two stages: first, we train an encoder that deterministically maps 3D assets into the parameters of an implicit function; second, we train a conditional diffusion model on outputs of the encoder. When trained on a large dataset of paired 3D and text data, our resulting models are capable of generating complex and diverse 3D assets in a matter of seconds. When compared to Point-E, an explicit generative model over point clouds, Shap-E converges faster and reaches comparable or better sample quality despite modeling a higher-dimensional, multi-representation output space. We release model weights, inference code, and samples at this https URL.

Released checkpoints

The authors released the following checkpoints:

• openai/shap-e: produces a 3D image from a text input prompt
• openai/shap-e-img2img: samples a 3D image from synthetic 2D image

Results

Image	Description
	A bird
	A shark
	A bowl of vegetables

Training details

Refer to the original paper.

Known limitations and potential biases

Refer to the original model card.

🔧 Technical Details

The paper provides detailed information about the model architecture, training process, and experimental results. For more technical details, please refer to Shap-E: Generating Conditional 3D Implicit Functions.

📄 License

This project is licensed under the MIT license.

📖 Citation

@misc{jun2023shape,
      title={Shap-E: Generating Conditional 3D Implicit Functions}, 
      author={Heewoo Jun and Alex Nichol},
      year={2023},
      eprint={2305.02463},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}