Results 1 to 13 of 13

Thread: Google Brain fills in the details of very lo-res images

  1. #1
    HEXUS.admin
    Join Date
    Apr 2005
    Posts
    25,126
    Thanks
    0
    Thanked
    1,354 times in 512 posts

    Google Brain fills in the details of very lo-res images

    CSI style unpixelating system uses neural net and learned conditioning to clear things up.
    Read more.

  2. #2
    Now 100% Apple free cheesemp's Avatar
    Join Date
    Apr 2007
    Location
    Near the New forest
    Posts
    1,960
    Thanks
    99
    Thanked
    113 times in 75 posts
    • cheesemp's system
      • Motherboard:
      • Gigabyte Z77-D3H
      • CPU:
      • Intel i5 3570k @ 4.2
      • Memory:
      • 16gb Crucial Ballastix Elite DDR3 @1866
      • Storage:
      • 64Gb M4 + 240Gb arc100 + 500Gb HDD
      • Graphics card(s):
      • RX 480 8Gb Nitro+ OC
      • PSU:
      • Antec 650W Semi modular
      • Case:
      • Antec 300
      • Operating System:
      • Win 10
      • Monitor(s):
      • 23" Samsung LED
      • Internet:
      • FTTC ~30Mb

    Re: Google Brain fills in the details of very lo-res images

    So the CSI 'computer enhance' meme is starting to come true? That something I never expected. Wonder if this will help with crimes - You couldn't use it as evidence but could help with suspect wanted mug shots...
    Trust

    Laptop : Lenovo G505s A8-5550m 8Gb 240Gb SSD Radeon HD 8550G + Radeon HD 8570M dual graphics

  3. #3
    Senior Member
    Join Date
    Jun 2013
    Posts
    267
    Thanks
    40
    Thanked
    18 times in 15 posts

    Re: Google Brain fills in the details of very lo-res images

    Quote Originally Posted by cheesemp View Post
    So the CSI 'computer enhance' meme is starting to come true? That something I never expected. Wonder if this will help with crimes - You couldn't use it as evidence but could help with suspect wanted mug shots...
    No. It creates very "Hollywood" faces because it's creating believable composites from overlapping images in a database that match the angle. If the person looked like the image, it would be chance (and a relatively low chance, given that your average Joe/Jo is not photogenic).

    In the same way, look at the images with involve windows. It's smart enough to track subtle gradation of light from right to left across an image, pick up a frame shape, and decide that this thing on the right is: a light source; a gentle enough light source to be a window; a size and shape consistent with a window. But in one of Google's shots, the window is a skylight cut into the slanting roof of an attic room. To replace the window, the software has to insert a composite window-ish element, and it has vertical frames like the ones in its database.

  4. #4
    root Member DanceswithUnix's Avatar
    Join Date
    Jan 2006
    Location
    In the middle of a core dump
    Posts
    7,208
    Thanks
    259
    Thanked
    658 times in 574 posts
    • DanceswithUnix's system
      • Motherboard:
      • M5A-97 EVO R2.0
      • CPU:
      • FX-8350
      • Memory:
      • 16GB ECC 1333
      • Storage:
      • 660GB Linux, 500GB Games (Win 10)
      • Graphics card(s):
      • Sapphire Nitro R9 380 4GB
      • PSU:
      • 650W Corsair TX
      • Case:
      • Antec 300
      • Operating System:
      • Fedora 24 + Win 10 Pro 64 (yuk)
      • Monitor(s):
      • Benq XL2730Z 1440p + Samsung 2343BW 2048x1152
      • Internet:
      • 80Mb/20Mb VDSL

    Re: Google Brain fills in the details of very lo-res images

    Quote Originally Posted by cheesemp View Post
    So the CSI 'computer enhance' meme is starting to come true? That something I never expected. Wonder if this will help with crimes - You couldn't use it as evidence but could help with suspect wanted mug shots...
    No, the information in the fuzzed out images is gone and cannot ever be restored. This tech could make for a cracking TV upscaler, but not evidence.

    The network was trained with images of celebs. Given a fuzzed input, it hallucinates a celeb based on the input. Given a picture of me, it would hallucinate the closest celeb like face it could.

    I look forward to TVs having a de-pixellation mode to make tamed tv footage rude again, it would work very well for that

  5. #5
    Senior Member
    Join Date
    Jul 2009
    Location
    West Sussex
    Posts
    667
    Thanks
    35
    Thanked
    78 times in 75 posts
    • kompukare's system
      • Motherboard:
      • Asus P8Z77-V LX
      • CPU:
      • Intel i5-3570K
      • Memory:
      • 2 x 4GB Crucial PC3-10700 (666MHz)
      • Storage:
      • Sandisk Extreme 120GB SSD | WDC 1TB Green | Samsung 1Tb Spinpoint
      • Graphics card(s):
      • Gigabyte Windforce HD 7950
      • PSU:
      • Corsair CX400 (Seasonic) or Seasonic SII-330
      • Case:
      • Sunbeam Freezing Storm
      • Operating System:
      • Windows 7-64
      • Monitor(s):
      • NEC 2170NX (1600x1200, PVA)

    Re: Google Brain fills in the details of very lo-res images

    Quote Originally Posted by DanceswithUnix View Post
    The network was trained with images of celebs. Given a fuzzed input, it hallucinates a celeb based on the input. Given a picture of me, it would hallucinate the closest celeb like face it could.
    So cameo appearances are going to soar and celebs themselves will begin to wonder if they had accidentally wandered on the set of some film without knowing it?

  6. #6
    Now 100% Apple free cheesemp's Avatar
    Join Date
    Apr 2007
    Location
    Near the New forest
    Posts
    1,960
    Thanks
    99
    Thanked
    113 times in 75 posts
    • cheesemp's system
      • Motherboard:
      • Gigabyte Z77-D3H
      • CPU:
      • Intel i5 3570k @ 4.2
      • Memory:
      • 16gb Crucial Ballastix Elite DDR3 @1866
      • Storage:
      • 64Gb M4 + 240Gb arc100 + 500Gb HDD
      • Graphics card(s):
      • RX 480 8Gb Nitro+ OC
      • PSU:
      • Antec 650W Semi modular
      • Case:
      • Antec 300
      • Operating System:
      • Win 10
      • Monitor(s):
      • 23" Samsung LED
      • Internet:
      • FTTC ~30Mb

    Re: Google Brain fills in the details of very lo-res images

    Quote Originally Posted by DanceswithUnix View Post
    No, the information in the fuzzed out images is gone and cannot ever be restored. This tech could make for a cracking TV upscaler, but not evidence.

    The network was trained with images of celebs. Given a fuzzed input, it hallucinates a celeb based on the input. Given a picture of me, it would hallucinate the closest celeb like face it could.

    I look forward to TVs having a de-pixellation mode to make tamed tv footage rude again, it would work very well for that
    Um - Thats why I stated not for evidence but for mug shots. Sure it might be just a very generic face in the future but its better than some useless CCTV police often show now when try to find a wanted person.

    Edit: Also don't forget this is version 1 of this tech. I'm sure it will improve to support more generic faces.
    Trust

    Laptop : Lenovo G505s A8-5550m 8Gb 240Gb SSD Radeon HD 8550G + Radeon HD 8570M dual graphics

  7. #7
    Registered+
    Join Date
    Sep 2014
    Posts
    56
    Thanks
    0
    Thanked
    6 times in 4 posts

    Re: Google Brain fills in the details of very lo-res images

    The starting image was a bowl of fruit.

  8. Received thanks from:

    DanceswithUnix (09-02-2017)

  9. #8
    root Member DanceswithUnix's Avatar
    Join Date
    Jan 2006
    Location
    In the middle of a core dump
    Posts
    7,208
    Thanks
    259
    Thanked
    658 times in 574 posts
    • DanceswithUnix's system
      • Motherboard:
      • M5A-97 EVO R2.0
      • CPU:
      • FX-8350
      • Memory:
      • 16GB ECC 1333
      • Storage:
      • 660GB Linux, 500GB Games (Win 10)
      • Graphics card(s):
      • Sapphire Nitro R9 380 4GB
      • PSU:
      • 650W Corsair TX
      • Case:
      • Antec 300
      • Operating System:
      • Fedora 24 + Win 10 Pro 64 (yuk)
      • Monitor(s):
      • Benq XL2730Z 1440p + Samsung 2343BW 2048x1152
      • Internet:
      • 80Mb/20Mb VDSL

    Re: Google Brain fills in the details of very lo-res images

    Quote Originally Posted by cheesemp View Post
    Edit: Also don't forget this is version 1 of this tech. I'm sure it will improve to support more generic faces.
    The information is gone, all the neural net can do is fill in the gaps with generic features best guessed from colour matching the blobs. Give it a heavily pixellated bowl of fruit, and it would get out the best face it could match.

    Don't think of this as image processing, think of it as an interpretive painting. The net sees the blobs and decides which brush strokes to use to make something that looks like a face and hair. It is clever, but it can only ever be an educated guess.

  10. #9
    . bledd's Avatar
    Join Date
    Jul 2003
    Posts
    1,836
    Thanks
    22
    Thanked
    133 times in 83 posts

    Re: Google Brain fills in the details of very lo-res images

    Is that Gavin Rossdale from Bush?

  11. #10
    Now 100% Apple free cheesemp's Avatar
    Join Date
    Apr 2007
    Location
    Near the New forest
    Posts
    1,960
    Thanks
    99
    Thanked
    113 times in 75 posts
    • cheesemp's system
      • Motherboard:
      • Gigabyte Z77-D3H
      • CPU:
      • Intel i5 3570k @ 4.2
      • Memory:
      • 16gb Crucial Ballastix Elite DDR3 @1866
      • Storage:
      • 64Gb M4 + 240Gb arc100 + 500Gb HDD
      • Graphics card(s):
      • RX 480 8Gb Nitro+ OC
      • PSU:
      • Antec 650W Semi modular
      • Case:
      • Antec 300
      • Operating System:
      • Win 10
      • Monitor(s):
      • 23" Samsung LED
      • Internet:
      • FTTC ~30Mb

    Re: Google Brain fills in the details of very lo-res images

    Quote Originally Posted by DanceswithUnix View Post
    The information is gone, all the neural net can do is fill in the gaps with generic features best guessed from colour matching the blobs. Give it a heavily pixellated bowl of fruit, and it would get out the best face it could match.

    Don't think of this as image processing, think of it as an interpretive painting. The net sees the blobs and decides which brush strokes to use to make something that looks like a face and hair. It is clever, but it can only ever be an educated guess.
    I understand what you are saying. I just trying to point out even if it is a closest match face it will be more helpful than a heavily pixelated picture narrowing down suspects. After all show someone the 8x8 pixel picture and they won't have a clue who it is. This might just help a human think 'that looks a little bit like John Smith and he has clothes like that' - I'm not saying it a miracle - just another tool similar to photofits.
    Trust

    Laptop : Lenovo G505s A8-5550m 8Gb 240Gb SSD Radeon HD 8550G + Radeon HD 8570M dual graphics

  12. #11
    Admin Saracen's Avatar
    Join Date
    Jul 2003
    Posts
    16,776
    Thanks
    726
    Thanked
    2,844 times in 2,008 posts

    Re: Google Brain fills in the details of very lo-res images

    Quote Originally Posted by DanceswithUnix View Post
    ....

    I look forward to TVs having a de-pixellation mode to make tamed tv footage rude again, it would work very well for that
    Now that would, I suspect, reveal a very large ..... market.

    Noli nothis permittere te terere.


  13. #12
    Le Adder Noir CK_1985's Avatar
    Join Date
    Apr 2008
    Posts
    578
    Thanks
    35
    Thanked
    29 times in 24 posts
    • CK_1985's system
      • Motherboard:
      • Gigabyte GA-H87N-WiFi mini-ITX
      • CPU:
      • i5-4670K
      • Memory:
      • 8GB Corsair Vengeance 1600MHz
      • Storage:
      • 500GB Samsung 850EVO
      • Graphics card(s):
      • GeForce GTX660ti
      • PSU:
      • Silverstone Strider 450W SFX
      • Case:
      • Silverstone Sugo SG-05
      • Operating System:
      • Windows 10 Pro
      • Monitor(s):
      • Dell U2414H
      • Internet:
      • BT Infinity Fibre

    Re: Google Brain fills in the details of very lo-res images

    This is realy interesting tech. I agree with the various comments about it creating very generic 'hollywood' faces, which is of limited use. However I an see benefits:

    1. The examples given are starting from an extreme (8x8) and so it's not surprising they're extremely generic. If you're talking about the CSI-type use case, it would be interesting to see how well it performs on a 16x16 image, or a 32x32 one... If it can take a slightly blurry picture and make it much less blurry that could be much more reliable.
    2. Even in the extreme examples given, you could still use it in a CSI-style use case. Obviously not for a conviction, but for producing a photofit-type image that could help you narrow the search. With enough development, you could even quantify something like 'there is a 90% certainty that this is a reasonable likeness' - which could definitely hav value. Even DNA evidence only gives you a certain % of certainty.
    "I want to be young and wild, then I want to be middle aged and rich, then I want to be old and annoy people by pretending that I'm deaf..."

    my Hexus.Trust

  14. #13
    Senior Member
    Join Date
    May 2014
    Posts
    429
    Thanks
    79
    Thanked
    13 times in 13 posts
    • Xlucine's system
      • Motherboard:
      • Gigabyte Z97MX Gaming 5
      • CPU:
      • i5 4690K @stock, until I upgrade my graphics
      • Memory:
      • 8GB @2133 11-11-11-27
      • Storage:
      • Crucial MX500 512GB, 2TB storage
      • Graphics card(s):
      • Gigabyte 750ti OC
      • PSU:
      • Seasonic S12G-550
      • Case:
      • Silverstone TJ08-E

    Re: Google Brain fills in the details of very lo-res images

    Look at the top row - the two photos look like very different people. Courts are terrible at evaluating probability, this will be taken to be infallible. Any tool using this approach will be biased to output the training data, if you use this in court it's guaranteeing a miscarriage of justice.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •