It is currently 16 Apr 2024, 08:06
   
Text Size

Upscale low res card images to high res using Deep Learning

Discuss Card Scans and Other Artwork Here

Moderator: CCGHQ Admins

Re: Upscale low res card images to high res using Deep Learn

Postby Huggybaby » 21 Sep 2017, 21:38

This project is amazing! These images are light years ahead of what we had in the old days.

austinio7116, please update your first post with a current before and after shot. Also leave up the pics you have now so we can see the progress you've achieved.
User avatar
Huggybaby
Administrator
 
Posts: 3205
Joined: 15 Jan 2006, 19:44
Location: Finally out of Atlanta
Has thanked: 696 times
Been thanked: 594 times

Re: Upscale low res card images to high res using Deep Learn

Postby MaxMtg » 22 Sep 2017, 10:19

I tried to use Waifu 2x to better the image quality of CMA low-res pictures but they came vastly inferior to your image quality
MaxMtg
 
Posts: 102
Joined: 11 Aug 2017, 16:11
Has thanked: 50 times
Been thanked: 6 times

Re: Upscale low res card images to high res using Deep Learn

Postby odino182 » 22 Sep 2017, 13:06

austinio7116 can you this amazing work for "stardand" italian card? ty alot
odino182
 
Posts: 42
Joined: 25 Jun 2008, 13:11
Has thanked: 3 times
Been thanked: 1 time

Re: Upscale low res card images to high res using Deep Learn

Postby austinio7116 » 20 Oct 2017, 20:02

MaxMtg wrote:I tried to use Waifu 2x to better the image quality of CMA low-res pictures but they came vastly inferior to your image quality
Yes, I think my final results are better for magic cards than Waifu, although I do think Waifu is great for anime art.
User avatar
austinio7116
 
Posts: 451
Joined: 10 Mar 2017, 11:59
Has thanked: 47 times
Been thanked: 169 times

Re: Upscale low res card images to high res using Deep Learn

Postby austinio7116 » 22 Oct 2017, 06:16

Here's the low res and upsampled versions of one of the nicer results on the art from XLN:

Emperor's Vanguard.full (2).jpg

Emperor's Vanguard.full.png.xhlq (2).jpg
User avatar
austinio7116
 
Posts: 451
Joined: 10 Mar 2017, 11:59
Has thanked: 47 times
Been thanked: 169 times

Re: Upscale low res card images to high res using Deep Learn

Postby austinio7116 » 27 Oct 2017, 22:21

Here are some sets of images showing the low res, the super resolution - and now the newly released scans for comparison:

compare1.jpg

compare2.jpg

compare3.jpg

compare4.jpg
User avatar
austinio7116
 
Posts: 451
Joined: 10 Mar 2017, 11:59
Has thanked: 47 times
Been thanked: 169 times

Re: Upscale low res card images to high res using Deep Learn

Postby kevlahnota » 28 Oct 2017, 02:53

Nice comparison. IMHO, if you are using mobile device not greater than 9" display, the super resolution will do the job (also the file size is very friendly for mobile devices). The card scans is much suited to PC with big monitors...
User avatar
kevlahnota
Programmer
 
Posts: 825
Joined: 19 Jul 2010, 17:45
Location: Philippines
Has thanked: 14 times
Been thanked: 264 times

Re: Upscale low res card images to high res using Deep Learn

Postby austinio7116 » 28 Oct 2017, 12:34

Here is a comparison with other upsampling methods:

The pictures are: Low Res, Bicubic interpolation, Waifu 2x (just the standard web version run twice - not specially trained on Magic cards), My DCGAN, HQ scan

compare5-1.jpg
User avatar
austinio7116
 
Posts: 451
Joined: 10 Mar 2017, 11:59
Has thanked: 47 times
Been thanked: 169 times

Re: Upscale low res card images to high res using Deep Learn

Postby Ayssia » 31 Oct 2017, 02:20

Any Iconic Masters update?

Edit: nvm... found it, sry.
Ayssia
 
Posts: 18
Joined: 14 Feb 2013, 23:21
Has thanked: 12 times
Been thanked: 3 times

Re: Upscale low res card images to high res using Deep Learn

Postby MaxMtg » 02 Nov 2017, 11:14

Can you upscale Duel Decks goblins vs merfolk?
MaxMtg
 
Posts: 102
Joined: 11 Aug 2017, 16:11
Has thanked: 50 times
Been thanked: 6 times

Re: Upscale low res card images to high res using Deep Learn

Postby austinio7116 » 02 Nov 2017, 11:27

Yes, I'll have a look at those.
User avatar
austinio7116
 
Posts: 451
Joined: 10 Mar 2017, 11:59
Has thanked: 47 times
Been thanked: 169 times

Re: Upscale low res card images to high res using Deep Learn

Postby austinio7116 » 02 Nov 2017, 22:58

User avatar
austinio7116
 
Posts: 451
Joined: 10 Mar 2017, 11:59
Has thanked: 47 times
Been thanked: 169 times

Re: Upscale low res card images to high res using Deep Learn

Postby MaxMtg » 03 Nov 2017, 00:06

austinio7116 wrote:https://drive.google.com/drive/folders/1gnO6dR8Tjlu3IhoxIMu5kvCeMzEYaGbL?usp=sharing

lowres


highres
It's amazing how those sometimes look even better than the scanned cards
Have you done this to commander anthology?
MaxMtg
 
Posts: 102
Joined: 11 Aug 2017, 16:11
Has thanked: 50 times
Been thanked: 6 times

Re: Upscale low res card images to high res using Deep Learn

Postby austinio7116 » 03 Nov 2017, 06:31

I think the colour depth, vibrance and colour consistency can be better with this method as printing/scanning is bound to reduce this a little, however the detail and sharpness can never be as good as the scans at 300dpi.

If I can find 370x265 CMA images from Wizards I should be able to upsample them.
User avatar
austinio7116
 
Posts: 451
Joined: 10 Mar 2017, 11:59
Has thanked: 47 times
Been thanked: 169 times

Re: Upscale low res card images to high res using Deep Learn

Postby austinio7116 » 03 Nov 2017, 09:08

I can't see 370x265 images, only lower Res gatherer ones, perhaps that's why your Waifu upsampling was so noticeably inferior. I can try upsampling the gatherer images but would not expect to get such good results, the network is specifically trained on, and so optimised for 370x265 input.
User avatar
austinio7116
 
Posts: 451
Joined: 10 Mar 2017, 11:59
Has thanked: 47 times
Been thanked: 169 times

PreviousNext

Return to Pictures

Who is online

Users browsing this forum: No registered users and 35 guests


Who is online

In total there are 35 users online :: 0 registered, 0 hidden and 35 guests (based on users active over the past 10 minutes)
Most users ever online was 4143 on 23 Jan 2024, 08:21

Users browsing this forum: No registered users and 35 guests

Login Form