🧪 Test?View on arXiv
Synthetic Leprosy Image Generation Using Mask-Conditioned Latent Diffusion and Transfer Learning from Large Chronic Wound Datasets
Not specified in the provided content
fine-tuninglatent diffusionimage synthesis
2609.13226
Builder Relevance
2h ago70%
Abstract
This paper explores the use of a model trained on chronic wound photography to generate synthetic leprosy images in a low-data regime.
Reality Card
Core Claim
Chronic wound photography is a viable donor domain for leprosy lesion synthesis, achieving perceptual diversity in generated images comparable to real leprosy images.
Method / Result
The generated set shows an internal perceptual diversity of 0.662, statistically indistinguishable from the real leprosy set's diversity of 0.672.
Limitations
The main limitation is the semantic control over the generated images rather than image quality.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.