'

Nvidia's AI-driven photo reconstruction imaging achieves surprisingly realistic results

Nvidia can use AI to reconstruct corrupted or edited images

Nvidia's AI-driven photo reconstruction imaging achieves surprisingly realistic results

Nvidia's AI-driven photo reconstruction imaging achieves surprisingly realistic results

It should be well known by now that Nvidia loves AI, with deep learning acting as a significant growth vector for the company thanks to the performance offered by their graphics hardware and their software SDKs. 

The company has now released a video which showcases what some of their research and development has yielded, using deep learning to create a tool which can restore detail to corrupted images or grant users the ability to delete data from images and fill them with seemingly accurate visual information. 

In one of the examples in the video below, Nvidia removed a small ground flag from an image, with their image restoration tool filling in the gaps to create an image which showcases the ground as if the flag never existed in the original picture.   

This tool can create shockingly realistic images, even when applied to areas of a human face, though reconstructed eyes seem to deliver an uncanny valley appearance, which is perhaps a good thing for now. 
 

       Researchers from NVIDIA, led by Guilin Liu, introduced a state-of-the-art deep learning method that can edit images or reconstruct a corrupted image, one that has holes or is missing pixels. The method can also be used to edit images by removing content and filling in the resulting holes. Learn more about their research paper “Image Inpainting for Irregular Holes Using Partial Convolutions”

 

This application of Deep learning is one of many potential uses of AI in software, allowing complex algorithms to be developed that can create detail in images that are otherwise missing or even upscale or "enhance" images with reasonable levels of detail. 

You can join the discussion on Nvidia's image detail reconstruction techniques on the OC3D Forums.  

«Prev 1 Next»

Most Recent Comments

23-04-2018, 17:53:20

Dicehunter
That's pretty amazing !Quote

23-04-2018, 18:21:17

NeverBackDown
Man would like to know what kind of math was involved with this. AI tends to heavily use linear algebra but something like this seems much more complex.Quote
Reply
x

Register for the OC3D Newsletter

Subscribing to the OC3D newsletter will keep you up-to-date on the latest technology reviews, competitions and goings-on at Overclock3D. We won't share your email address with ANYONE, and we will only email you with updates on site news, reviews, and competitions and you can unsubscribe easily at any time.

Simply enter your name and email address into the box below and be sure to click on the links in the confirmation emails that will arrive in your e-mail shortly after to complete the registration.

If you run into any problems, just drop us a message on the forums.