StableNormal

Forked from https://huggingface.co/spaces/Stable-X/StableNormal.

This is the modified fork version of the StableNormal gradio space for EasyVolcap format dataset processing.

Installation

To install the required dependencies, you can use the following command:

sh setup.sh

Usage

Once you have a prepared EasyVolcap format dataset, and have EasyVolcap installed, you can use the following command to compute the monocular normal maps for the dataset:

python run.py \
    --data_root /path/to/dataset/root \
    --resolusion 1024 \
    --align 64 \
    --scenes /list/of/scene/names

• data_root: The root directory of the dataset, including a list of scenes inside.
• resolusion: The network processing resolution, default to 1024.
• align: The network alignment resolution, default to 64.
• scenes: This argument is optional, if not provided, all scenes in the dataset will be processed, otherwise only the specified scenes will be processed, split by space.

Notes

• We use Stable-X/yoso-normal-v0-3 and Stable-X/stable-normal-v0-1 for now.
• Resolution issue: https://github.com/Stable-X/StableNormal/issues/31