supersambat/ffhqdat-4x-upscaler

4x face image upscaler trained on FFHQ dataset using DAT (Dual Aggregation Transformer) architecture. Optimized for portrait and face photos.

Public
19 runs
License
Weights
Paper
Playground API Examples README Versions

Examples

View more

Run time and cost

This model runs on Nvidia L40S GPU hardware. We don't yet have enough runs of this model to provide performance information.

Readme

4x-FFHQDAT Face Upscaler

A 4x super-resolution model specifically trained for face/portrait images using the FFHQ dataset.

Model Details

  • Architecture: DAT (Dual Aggregation Transformer)
  • Scale: 4x upscaling
  • Training data: FFHQ (259,990 face images)
  • Best for: Portrait and face photos
  • License: CC-BY-4.0 — credit original author Helaman

Usage

Input a face image, get a 4x upscaled version back.

Credits

Original model by Helaman on OpenModelDB.

Model created
Model updated