SAM (Style-based Age Manipulation) Model

This is the SAM model for realistic face aging/de-aging transformations. Used in the DermaIntel application for the Aging Simulation feature.

Model Details

  • Architecture: StyleGAN2-based age manipulation
  • Training Data: FFHQ (Flickr-Faces-HQ)
  • File: sam_ffhq_aging.pt (~2.1 GB)
  • Purpose: Age transformation (aging and de-aging faces)

Usage

from huggingface_hub import hf_hub_download

# Download model
model_path = hf_hub_download(
    repo_id="PayamFard123/dermaintel-sam-aging",
    filename="sam_ffhq_aging.pt"
)

# Load with PyTorch
import torch
checkpoint = torch.load(model_path, map_location='cpu')

Features

  • Realistic aging effects (wrinkles, skin texture changes)
  • De-aging (smoothing, rejuvenation)
  • Preserves facial identity during transformation
  • Works with various age ranges (10-90 years)

Original Source

Based on SAM (Style-based Age Manipulation) research.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support