Skip to content
/ LEAF Public

[MICCAI 2025] LEAF: Latent Diffusion with Efficient Encoder Distillation for Aligned Features in Medical Image Segmentation

License

Notifications You must be signed in to change notification settings

Pearisli/LEAF

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

leaf LEAF: Latent Diffusion with Efficient Encoder Distillation for Aligned Features in Medical Image Segmentation

Page Paper HuggingFace Model

LEAF Framework

leaf_pipeline

Performance

leaf_performance

Ablation

leaf_ablation

Update

Setup

  1. Clone the repository:
git clone https://github.com/lispear/LEAF.git
cd LEAF-master
  1. Install dependencies (requires conda):
conda create -n leaf python=3.11.11 -y
conda activate leaf
pip install -r requirements.txt 

Training

  1. Create assets directory:
mkdir assets
cd assets
  1. Prepare pre-trained models:
  • Download U-Net and VAE and extract weights
unzip kl-f8.zip -d vae
unzip lsun_churches.zip -d unet
cd ..
python extract_weights.py
  1. Run training script:
accelerate launch \
    --num_processes 1 \
    --num_machines 1 \
    --mixed_precision 'no' \
    --dynamo_backend 'no' \
    train.py --config config.yaml

About

[MICCAI 2025] LEAF: Latent Diffusion with Efficient Encoder Distillation for Aligned Features in Medical Image Segmentation

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages