Guided Lensless Polarization Imaging

Pretrained checkpoints for RGB-guided super-resolution of lensless polarimetric images.

Code: GitHub

Models

Checkpoint Architecture Guide Description
grayscale_with_guide.pth swinfusionSR_GRAYSCALE RGB 3ch grayscale polarimetric (0°, 45°, 90°) with RGB guide
grayscale_no_guide.pth swinfusionSR_GRAYSCALE None 3ch grayscale polarimetric, no guide (ablation)
color_with_guide.pth swinfusionSRcolor RGB 12ch color polarimetric (4 angles × 3 RGB) with RGB guide
color_no_guide.pth swinfusionSRcolor None 12ch color polarimetric, no guide (ablation)

Usage

git clone https://github.com/noakraicertau/Guided-Lensless-Polarization-Imaging.git
cd Guided-Lensless-Polarization-Imaging
# Download a checkpoint
wget https://huggingface.co/noakraicer/Guided-Lensless-Polarization-Imaging/resolve/main/grayscale_with_guide.pth
# Set path/pretrained_netG in your config JSON, then:
python test_swin_final.py --opt options/test_RGBP.json
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