Image-Text-to-Image
Safetensors
English
SPAD
Photons
Generative
ISP

gQIR: Generative Quanta Image Reconstruction

Aryan Garg1, Sizhuo Ma2, Mohit Gupta1

1 University of Wisconsin-Madison 2 Snap, Inc

color_spads

All model weights are available here now!

Color-Model Name Stage Bit Depth πŸ€— Download Link
qVAE Stage 1 1-bit 1965000.pt
Adversarial Diffusion LoRA-UNet Stage 2 1-bit state_dict.pth
qVAE Stage 1 3-bit 0105000.pt
Adversarial Diffusion LoRA-UNet Stage 2 3-bit state_dict.pth
FusionViT Stage 3 3-bit fusion_vit_0050000.pt
Monochrome-Model Stage Bit Depth πŸ€— Download Link
qVAE Stage 1 3-bit 0150000.pt
Adversarial Diffusion LoRA-UNet Stage 2 3-bit state_dict.pth
FusionViT Stage 3 3-bit fusion_vit_0020000.pt

Code at github.com/Aryan-Garg/gQIR

ArXiv Version: arxiv.org/abs/2602.20417

Cite Us:

@InProceedings{garg_2026_gqir,
    author    = {Garg, Aryan and Ma, Sizhuo and  Gupta, Mohit},
    title     = {gQIR: Generative Quanta Image Reconstruction},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2026},
}
Downloads last month

-

Downloads are not tracked for this model. How to track
Safetensors
Model size
1B params
Tensor type
F32
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for aRy4n/gQIR

Finetuned
(1)
this model

Datasets used to train aRy4n/gQIR

Spaces using aRy4n/gQIR 2

Paper for aRy4n/gQIR