NVIDIA Develops AI That Can Remove Noise, Grain, And Even Watermarks From Photos

Researchers from NVIDIA, Aalto University, and MIT have developed an AI that can remove noise from grainy photos and automatically enhance them. This technology can be beneficial in several real-world situations where it is difficult to obtain clear image data like MRI scans, astronomical imaging, and more.

Existing noise-reduction AI systems require both noisy and clean input images, but NVIDIA’s AI can restore images without being shown what the noise-free image looks like. It just needs to look at examples of corrupted images. The researchers trained the AI on 50,000 images and the deep-learning algorithm was able to produce impressive results. Check them out below.

1.

NVIDIA AI can remove grains, noise, and watermarks from photos: Before and after - 1

2.

NVIDIA AI can remove grains, noise, and watermarks from photos: Before and after - 2

3.

NVIDIA AI can remove grains, noise, and watermarks from photos: Before and after - 3

4.

NVIDIA AI can remove grains, noise, and watermarks from photos: Before and after - 4

5.

NVIDIA AI can remove grains, noise, and watermarks from photos: Before and after - 5

6. The AI can remove text as well

NVIDIA AI can remove grains, noise, and watermarks from photos: Before and after - 6

7.

NVIDIA AI can remove grains, noise, and watermarks from photos: Before and after - 7

8.

NVIDIA AI can remove grains, noise, and watermarks from photos: Before and after - 8

Watch the AI in action below

00:11 – Train a denoiser without clean targets
00:24 – Poisson Noise
00:34 – Gaussian Noise
00:43 – Monte Carlo Rendering
00:48 – Impulse Noise
00:53 – MRI
00:57 – Text Removal

Read the research paper →

What do you think of NVIDIA’s new technology? Share this post with a designer or a photographer and voice your views in the comments below. Source: PetaPixel

Don’t Miss:

Fuente; https://digitalsynopsis.com/design/nvidia-ai-removes-noise-grain-watermarks-from-photos/

0 comentarios

Dejar un comentario

¿Quieres unirte a la conversación?
Siéntete libre de contribuir

Deja un comentario

Tu dirección de correo electrónico no será publicada. Los campos obligatorios están marcados con *