Addlink S70 1TB NVMe M.2 SSD review
G.Skill KM360 Keyboard Review
Promo: VIP-SCDKey Games and Windows 10 / Office
Alphacool EISBAER Ex BLACK LCS review
Netgear Nighthawk AX12 Wi-Fi 6 router review
Corsair HS60 PRO Surround Headset Review
MSI Radeon RX 5700 Gaming X review
Team Group Cardea II NVMe 1 TB SSD Review
ASRock X570 Phantom Gaming ITX/TB3 review
Promo: URCDKey Autumn Sale Windows 10 Pro for $11
Nvidia: AI Reconstructs Photos with Realistic Results
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. This is pretty cool to watch actually.
NVIDIA: GPU prices will continue increasing through Q3 2018 - 02/25/2018 11:39 AM
You guys all have been able to observe the graphics card price spikes the past half year. Though things are a bit more steady and stable it seems that the trend might continue for a while. Nvidia now...
NVIDIA: DirectX 12 Is Huge Improvement for Gaming - 03/21/2014 08:12 AM
Microsoft today introduced DirectX 12 at the annual Game Developers Conference (GDC) in San Francisco. DX12 is Microsoft's latest version of the graphics API that is the dominant standard in the grow...
NVIDIA: New consoles will never match PC graphics - 09/26/2013 04:24 PM
NVIDIA Senior Vice President of Content and Technology Tony Tamasi mentioned in an interview that game consoles leapfrogging PC graphic capabilities are definitely a thing of the past. Tamasi said Mic...
NVIDIA: Era of fully-custom console hardware is over - 01/18/2013 06:28 PM
NVIDIA Ceo Jen-Hsun Huang states that that fully-custom hardware for video game consoles is no longer feasible. Huang explains that with the power delivered by today's CPUs and GPUs, it no longer mak...
NVIDIA: Project Denver is ARMv8 with added secret sauce - 11/12/2012 09:59 AM
In a conference call last week with financial analysts, NVIDIA CEO Jen-Hsun Huang revealed that the company's Project Denver, aka NVIDIA's upcoming desktop processor, is a full-custom ARMv8 archite...
Pawel04
Member
Posts: 66
Joined: 2015-11-19
Member
Posts: 66
Joined: 2015-11-19
#5540340 Posted on: 04/24/2018 09:43 AM
Would agree with you there, i think humans still do better than this for the time being.
I mean, this looks good on the small scale, but I don't think they're yet there to have it done in a quality that couldn't be identified as "fake"
Would agree with you there, i think humans still do better than this for the time being.
bipman
Junior Member
Posts: 5
Joined: 2015-12-20
Junior Member
Posts: 5
Joined: 2015-12-20
#5540345 Posted on: 04/24/2018 09:55 AM
It depends what is the goal and how much resource you can spend.
Human can do better, of course, but takes much more time or costs more.
e.g. removing watermarks or an electric wire / pole from a beautyful panorama view a removing a funny photobomber takes only a few seconds with this tool, but manually can take hours for an amateur and even long minutes for a pro.
And for an amateur or facebook/instagram user time is more important than quality.
It depends what is the goal and how much resource you can spend.
Human can do better, of course, but takes much more time or costs more.
e.g. removing watermarks or an electric wire / pole from a beautyful panorama view a removing a funny photobomber takes only a few seconds with this tool, but manually can take hours for an amateur and even long minutes for a pro.
And for an amateur or facebook/instagram user time is more important than quality.
Kaarme
Senior Member
Posts: 1622
Joined: 2013-03-10
Senior Member
Posts: 1622
Joined: 2013-03-10
#5540347 Posted on: 04/24/2018 09:58 AM
That was decently impressive, but I'd say only assuming one could do that at home with their regular Nvidia video card and without needing to pay hundreds, if not thousands, of dollars for the software. Of course it being a little slower would be acceptable, depending on the GPU. If the software was published only to professionals working for big companies, like media houses, and they needed to pay big money to use it for a year... Nah, I wouldn't call it impressive anymore.
That was decently impressive, but I'd say only assuming one could do that at home with their regular Nvidia video card and without needing to pay hundreds, if not thousands, of dollars for the software. Of course it being a little slower would be acceptable, depending on the GPU. If the software was published only to professionals working for big companies, like media houses, and they needed to pay big money to use it for a year... Nah, I wouldn't call it impressive anymore.
FM57
Senior Member
Posts: 156
Joined: 2016-12-28
Senior Member
Posts: 156
Joined: 2016-12-28
#5540348 Posted on: 04/24/2018 10:01 AM
Whaooouhh. I definitely know now who to contact to get a younger and nicer picture of me, myself and I for my LinkedIn profile.
Whaooouhh. I definitely know now who to contact to get a younger and nicer picture of me, myself and I for my LinkedIn profile.
Fox2232
Senior Member
Posts: 9738
Joined: 2012-07-20
Senior Member
Posts: 9738
Joined: 2012-07-20
#5540396 Posted on: 04/24/2018 12:33 PM
Better, but nowhere near as fast.
Would agree with you there, i think humans still do better than this for the time being.
Better, but nowhere near as fast.
lucasfrance
Member
Posts: 33
Joined: 2017-04-07
Member
Posts: 33
Joined: 2017-04-07
#5540397 Posted on: 04/24/2018 12:33 PM
Check https://www.theinpaint.com/ the Inpaint sw does this for a long time alsready !
Check https://www.theinpaint.com/ the Inpaint sw does this for a long time alsready !
Denial
Senior Member
Posts: 12342
Joined: 2004-05-16
Senior Member
Posts: 12342
Joined: 2004-05-16
#5540417 Posted on: 04/24/2018 01:13 PM
Inpaint and tools like content aware fill in photoshop sample the surrounding area and just paste it over the target you're trying to remove. It works for most things but you can see the patterns it's sampling from and won't work in huge areas. The technique Nvidia is using actually generates imagery based on what's being targeted from a library of over 20K pictures. If you used inpaint to replace the eyes on a person, or photoshop, it wouldn't generate new eyes like this does.
It's a similar output but how the technique is being performed is way different. With a larger dataset and more training the Nvidia technique will be far more accurate and less prone to 'patterning"
Check https://www.theinpaint.com/ the Inpaint sw does this for a long time alsready !
Inpaint and tools like content aware fill in photoshop sample the surrounding area and just paste it over the target you're trying to remove. It works for most things but you can see the patterns it's sampling from and won't work in huge areas. The technique Nvidia is using actually generates imagery based on what's being targeted from a library of over 20K pictures. If you used inpaint to replace the eyes on a person, or photoshop, it wouldn't generate new eyes like this does.
It's a similar output but how the technique is being performed is way different. With a larger dataset and more training the Nvidia technique will be far more accurate and less prone to 'patterning"
lucasfrance
Member
Posts: 33
Joined: 2017-04-07
Member
Posts: 33
Joined: 2017-04-07
#5540434 Posted on: 04/24/2018 02:23 PM
Inpaint and tools like content aware fill in photoshop sample the surrounding area and just paste it over the target you're trying to remove. It works for most things but you can see the patterns it's sampling from and won't work in huge areas. The technique Nvidia is using actually generates imagery based on what's being targeted from a library of over 20K pictures. If you used inpaint to replace the eyes on a person, or photoshop, it wouldn't generate new eyes like this does.
It's a similar output but how the technique is being performed is way different. With a larger dataset and more training the Nvidia technique will be far more accurate and less prone to 'patterning"
You're right! Inpaint is clearly not able to create "eyes" as in the Nvidia demo... But it works well for a picture small area cleaning.
Inpaint and tools like content aware fill in photoshop sample the surrounding area and just paste it over the target you're trying to remove. It works for most things but you can see the patterns it's sampling from and won't work in huge areas. The technique Nvidia is using actually generates imagery based on what's being targeted from a library of over 20K pictures. If you used inpaint to replace the eyes on a person, or photoshop, it wouldn't generate new eyes like this does.
It's a similar output but how the technique is being performed is way different. With a larger dataset and more training the Nvidia technique will be far more accurate and less prone to 'patterning"
You're right! Inpaint is clearly not able to create "eyes" as in the Nvidia demo... But it works well for a picture small area cleaning.
Stormyandcold
Senior Member
Posts: 5201
Joined: 2003-09-15
Senior Member
Posts: 5201
Joined: 2003-09-15
#5540439 Posted on: 04/24/2018 02:28 PM
I like it. For consumers, the quality is acceptable, especially for most of the scenery stuff. On humans, it's not as effective...yet.
I like it. For consumers, the quality is acceptable, especially for most of the scenery stuff. On humans, it's not as effective...yet.
schmidtbag
Senior Member
Posts: 4427
Joined: 2012-11-10
Senior Member
Posts: 4427
Joined: 2012-11-10
#5540449 Posted on: 04/24/2018 02:55 PM
I personally think this is really cool. As others pointed out, this isn't just simply making due with whatever textures are currently available in the photo. Though I found the eye replacements a little bit shocking (even a bit funny), I also found it to be pretty fascinating.
Furthermore, this was being run on a Linux system, which I personally like since (to my knowledge) there is no Linux-native equivalent to this kind of thing.
But seeing as this is Nvidia, I can't help but think this is going to use CUDA or some other proprietary tech. I would think this would mostly be a CPU-intensive task, or at least not all that beneficial of GPU use (regardless of brand).
I personally think this is really cool. As others pointed out, this isn't just simply making due with whatever textures are currently available in the photo. Though I found the eye replacements a little bit shocking (even a bit funny), I also found it to be pretty fascinating.
Furthermore, this was being run on a Linux system, which I personally like since (to my knowledge) there is no Linux-native equivalent to this kind of thing.
But seeing as this is Nvidia, I can't help but think this is going to use CUDA or some other proprietary tech. I would think this would mostly be a CPU-intensive task, or at least not all that beneficial of GPU use (regardless of brand).
-Tj-
Senior Member
Posts: 16385
Joined: 2012-05-18
Senior Member
Posts: 16385
Joined: 2012-05-18
#5540483 Posted on: 04/24/2018 04:38 PM
This Iooks perfect for my scenery editing pics
Where can I download this??
This Iooks perfect for my scenery editing pics
Where can I download this??
dragonlord
Senior Member
Posts: 133
Joined: 2005-02-17
Senior Member
Posts: 133
Joined: 2005-02-17
#5540544 Posted on: 04/24/2018 08:07 PM
Great for stills. Call me when it works on VIDEO, which is where the real challenges are in this space.
Great for stills. Call me when it works on VIDEO, which is where the real challenges are in this space.
Picolete
Senior Member
Posts: 270
Joined: 2014-12-09
Senior Member
Posts: 270
Joined: 2014-12-09
#5540568 Posted on: 04/24/2018 09:32 PM
It gives you anime eyes.
Wake me up when they reach the levels of human picture reconstruction
It gives you anime eyes.
Wake me up when they reach the levels of human picture reconstruction
bigfutus
Senior Member
Posts: 522
Joined: 2003-11-22
Senior Member
Posts: 522
Joined: 2003-11-22
#5540571 Posted on: 04/24/2018 09:36 PM
Content aware stuff in Photoshop worked always perfect in Adobe presentations too, but in real life sometimes it acts stupid, and it's much easier to stamp it manually.
Content aware stuff in Photoshop worked always perfect in Adobe presentations too, but in real life sometimes it acts stupid, and it's much easier to stamp it manually.
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 10738
Joined: 2014-07-21
I mean, this looks good on the small scale, but I don't think they're yet there to have it done in a quality that couldn't be identified as "fake"