Jeez, I’m clearly not up to date with proprietary denoising. That’s silly good results.
Proprietary software is “better” at ML powered tasks like denoising, upscaling and automasking, I guess. Nobody’s saying that can’t change though. 
1 Like
Which models did you try and how did you set up chaiNNer? It seems there’s no Pytorch or ONNX node for denoising, so right now I’m just using an upscale node to run the model.
Here’s the dataset that most of the denoising models seem to have been trained on: SIDD
There’s also a paper describing their methodology for building the dataset.
Seems that most public datasets are pretty bad:
Denoising benchmark with real images There have been,
to the best of our knowledge, two attempts to quantitatively
benchmark denoising algorithms on real images. One is the
RENOIR dataset [2], which contains pairs of low/high-ISO
images. This dataset lacks accurate spatial alignment, and
the low-ISO images still contain noticeable noise. Also, the
raw image intensities are linearly mapped to 8-bit depth,
which adversely affects the quality of the images.
More closely related to our effort is the work on the
Darmstadt Noise Dataset (DND) [25]. Like the RENOIR
dataset, DND contains pairs of low/high-ISO images. By
contrast, the work in [25] post-processes the low-ISO im-
ages to (1) spatially align them to their high-ISO counter-
parts, and (2) overcome intensity changes due to changes
in ambient light or artificial light flicker. This work was
the first principled attempt at producing high-quality ground
truth images. However, most of the DND images have rel-
atively low levels of noise and normal lighting conditions.
As a result, there is a limited number of cases of high noise
levels or low-light conditions, which are major concerns
for image denoising and computer vision in general. Also,
treating misalignment between images as a global transla-
tion is not sufficient for cases including lens motion, radial
distortion, or optical image stabilization.
If anyone is looking for a project, building a high-quality DSLR/mirrorless dataset sounds like a splendid idea 
1 Like
Same here, I used an upscale node. I’ll check the models in the evening, when I have more time. Some seemed to be ineffective (I could not discern any difference).
?
was hoping once @trougnouf is out of the phd defense strains we could use his dataset to improve the vkdt module @Tamas_Papp refered to above, vkdt/src/pipe/modules/jddcnn at master · hanatos/vkdt · GitHub
best with some pytorch-to-vkdt code compiler, and ideally including the great ideas that @David_Tschumperle showed in Machine Learning Library in G'MIC - #89 by David_Tschumperle .
6 Likes
“Great ideas” and “David Tschumperlé” in the same sentence. I never thought I’d read this in my lifetime 
6 Likes