English

Improved Generation of Synthetic Imaging Data Using Feature-Aligned Diffusion

Computer Vision and Pattern Recognition 2024-10-04 v1 Machine Learning

Abstract

Synthetic data generation is an important application of machine learning in the field of medical imaging. While existing approaches have successfully applied fine-tuned diffusion models for synthesizing medical images, we explore potential improvements to this pipeline through feature-aligned diffusion. Our approach aligns intermediate features of the diffusion model to the output features of an expert, and our preliminary findings show an improvement of 9% in generation accuracy and ~0.12 in SSIM diversity. Our approach is also synergistic with existing methods, and easily integrated into diffusion training pipelines for improvements. We make our code available at \url{https://github.com/lnairGT/Feature-Aligned-Diffusion}.

Keywords

Cite

@article{arxiv.2410.00731,
  title  = {Improved Generation of Synthetic Imaging Data Using Feature-Aligned Diffusion},
  author = {Lakshmi Nair},
  journal= {arXiv preprint arXiv:2410.00731},
  year   = {2024}
}

Comments

Accepted to First International Workshop on Vision-Language Models for Biomedical Applications (VLM4Bio 2024) at the 32nd ACM-Multimedia conference

R2 v1 2026-06-28T19:03:54.183Z