README.md

metadata

title: NormalCrafter
app_file: app.py
sdk: gradio
sdk_version: 5.23.3

NormalCrafter: Learning Temporally Consistent Video Normal from Video Diffusion Priors

Yanrui Bin¹,Wenbo Hu^2*, Haoyuan Wang³, Xinya Chen⁴, Bing Wang<sup>2 †

1</sup>Spatial Intelligence Group, The Hong Kong Polytechnic University ²ARC Lab, Tencent PCG ³City University of Hong Kong ⁴Huazhong University of Science and Technology

🔆 Notice

We recommend that everyone use English to communicate on issues, as this helps developers from around the world discuss, share experiences, and answer questions together.

For business licensing and other related inquiries, don't hesitate to contact [email protected].

🔆 Introduction

🤗 If you find NormalCrafter useful, please help ⭐ this repo, which is important to Open-Source projects. Thanks!

🔥 NormalCrafter can generate temporally consistent normal sequences with fine-grained details from open-world videos with arbitrary lengths.

• [24-04-01] 🔥🔥🔥 NormalCrafter is released now, have fun!

🚀 Quick Start

🤖 Gradio Demo

• Online demo: NormalCrafter
• Local demo:

  gradio app.py
  

🛠️ Installation

1. Clone this repo:

git clone [email protected]:Binyr/NormalCrafter.git

2. Install dependencies (please refer to requirements.txt):

pip install -r requirements.txt

🤗 Model Zoo

NormalCrafter is available in the Hugging Face Model Hub.

🏃‍♂️ Inference

1. High-resolution inference, requires a GPU with ~20GB memory for 1024x576 resolution:

python run.py  --video-path examples/example_01.mp4

2. Low-resolution inference requires a GPU with ~6GB memory for 512x256 resolution:

python run.py  --video-path examples/example_01.mp4 --max-res 512