Downloads · 30 days
0
low-light-denoise-4xsr
23f2003853/low-light-denoise-4xsr
low-light-denoise-4xsr is a machine learning model from 23f2003853. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
license: mit tags: - super-resolution - image-denoising - real-esrgan - pytorch ---
Access
Public
Updated Sep 6, 2026
Repo size
67 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.zip67 MB · 100%
At a glance
- Access
- Public
- Created
- Sep 6, 2026
- Updated
- Sep 6, 2026
- SHA
- 5b4e323c
Model card
From the Hugging Face model README
license: mit tags:
- super-resolution
- image-denoising
- real-esrgan
- pytorch
Low-Light Denoising + 4x Super-Resolution (Real-ESRGAN, fine-tuned)
Fine-tuned RRDBNet (Real-ESRGAN x4plus) generator for joint denoising and 4x
super-resolution of low-light images, trained on the Kaggle dlp-26t2-nppe3
competition dataset.
Usage
from basicsr.archs.rrdbnet_arch import RRDBNet
from huggingface_hub import hf_hub_download
import torch
path = hf_hub_download("{repo_id}", "finetuned_x4plus.pth")
model = RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=23, num_grow_ch=32, scale=4)
model.load_state_dict(torch.load(path, map_location="cpu")["params_ema"])
model.eval()
- Created
- Sep 6, 2026
- Updated
- Sep 6, 2026