OSU-NLP-Group/MagicBrush
A 10K-sample manually annotated dataset for training and evaluating instruction-guided image editing models using diffusion.
Not currently ranked — collecting fresh signals.
star history
MagicBrush is a large-scale dataset of image triplets (source, instruction, target) supporting instruction-guided image editing. It covers scenarios including single-turn, multi-turn, mask-provided, and mask-free editing. The repository includes training code and a fine-tuned checkpoint for the InstructPix2Pix model. The dataset is published on Huggingface and was accepted to NeurIPS 2023 Datasets and Benchmarks.
Frequently asked
- What is OSU-NLP-Group/MagicBrush?
- A 10K-sample manually annotated dataset for training and evaluating instruction-guided image editing models using diffusion.
- Is MagicBrush open source?
- Yes — OSU-NLP-Group/MagicBrush is an open-source project tracked on heatdrop.
- What language is MagicBrush written in?
- OSU-NLP-Group/MagicBrush is primarily written in Python.
- How popular is MagicBrush?
- OSU-NLP-Group/MagicBrush has 409 stars on GitHub.
- Where can I find MagicBrush?
- OSU-NLP-Group/MagicBrush is on GitHub at https://github.com/OSU-NLP-Group/MagicBrush.