Join the Unexplained Mysteries community today! It's free and setting up an account only takes a moment.
- Sign In or Create Account -
Sign in to follow this  
Followers 0
Moeyoatey

Using Ms paint to create realistic photos

12 posts in this topic

Since every photo is composed of pixels. I wonder how difficult it would be to either take a low definition photo and to fill in the gaps to make it high definition or take known photo using various set points to reproduce the photo from various angles. I tried to use a random number generator to assign various colors but it turned out to be a mess. The only other idea I had is use some type of fractal program but my programming skills are not that great. Any ideas?

Share this post


Link to post
Share on other sites

Fractals probably wouldnt do it because they're based off mathematical algorithms and just generate a graphical display of them. A program, apopysis, would be a good example of what you can do with fractals.

I'm not sure what you mean by angles but you would only be able to flip/rotate the image. Manipulating the image to

make it so it appears you have taken the photo standing up instead of sitting down I just don't see how it would be possible unless doing a painstaking edit of the entire photo or probably remake it altogether.

You also wouldnt be able to make a low res, hi res. simply because it would again be like making something from nothing again.

1 person likes this

Share this post


Link to post
Share on other sites

Posted (edited)

I think what you are asking for is only possible in Hollywood movies...

As Orcseeker says to make lo-res into hi-res you need to fill in the gaps. This can be done to a certain extent but as the gaps are filled in by 'guesses' what is created is a blur, a mixture of the surrounding elements. Just like when you enlarge a digital image. Going the other way is no problem (hi-res to lo-res) because you are throwing away information rather than making it up. I hope this makes sense. The programmers at Adobe etc and other companies have been refining this process for years, the results are improving (especially with blur removal etc) but IMO accurately turning lo-res to hi-res is a long way off if at all possible.

A better angle of attack might be using traditional film to capture the image you want to improve.

As for rotating a flat image, again this is impossible unless you have the image from that other angle. If you have a photo of the back of someone's head, how would a computer program know what the front of the face looked like? A Person might be able to create an interpretation of what they consider another view to be, software might be able to three dimensionally model some parts of the image and make your request seemingly possible but in reality again this is just guess work and not a reliable representation of what the camera cannot see.

What you are requesting pops up in TV and film occasionally but it is fantasy in reality.

As for the question in the title of your thread, you might like this...

https://www.youtube.... paint art work

Edited by Junior Chubb
2 people like this

Share this post


Link to post
Share on other sites

Since every photo is composed of pixels. I wonder how difficult it would be to either take a low definition photo and to fill in the gaps to make it high definition or take known photo using various set points to reproduce the photo from various angles. I tried to use a random number generator to assign various colors but it turned out to be a mess. The only other idea I had is use some type of fractal program but my programming skills are not that great. Any ideas?

Take decent software, highlight the picture and select enlarge.

Try GIMP 2 (free) or Photoshop (not free)

Share this post


Link to post
Share on other sites

Assuming you a have workspace that doesn't go on for infinity, there will be set number of pixels. Since each pixel can only be assigned a limited range of colors there is finite number of of combinations that can occur but it will be a huge number. Suppose you have a workspace of only 1 pixel, with 256 different colors that equals 256 different combinations. A workspace of 2 pixels would be 2^256 combinations. A high definition photo that is 1280 x 720 pixels would have 921600^256 different combinations. It is only a matter of being able to get every possible combination. If you were able to do that no only would you be able to do the things you asked but you would have essentially any photo that could ever exist.

2 people like this

Share this post


Link to post
Share on other sites

Posted (edited)

Assuming you a have workspace that doesn't go on for infinity, there will be set number of pixels. Since each pixel can only be assigned a limited range of colors there is finite number of of combinations that can occur but it will be a huge number. Suppose you have a workspace of only 1 pixel, with 256 different colors that equals 256 different combinations. A workspace of 2 pixels would be 2^256 combinations. A high definition photo that is 1280 x 720 pixels would have 921600^256 different combinations. It is only a matter of being able to get every possible combination. If you were able to do that no only would you be able to do the things you asked but you would have essentially any photo that could ever exist.

Did I misunderstand the question B?

He wants to turn a lo-def image into a hi-def image, and show the image (photo I would guess) from a different angle to which it was taken.

You can enlarge digital images and increase the DPI and classify it as hi-def, but in reality the loss in quality would make the image redundant regardless of how many colours are in use.

Edited by Junior Chubb

Share this post


Link to post
Share on other sites

Assuming you a have workspace that doesn't go on for infinity, there will be set number of pixels. Since each pixel can only be assigned a limited range of colors there is finite number of of combinations that can occur but it will be a huge number. Suppose you have a workspace of only 1 pixel, with 256 different colors that equals 256 different combinations. A workspace of 2 pixels would be 2^256 combinations. A high definition photo that is 1280 x 720 pixels would have 921600^256 different combinations. It is only a matter of being able to get every possible combination. If you were able to do that no only would you be able to do the things you asked but you would have essentially any photo that could ever exist.

That is what I was trying to get at. I wanted to some how manipulate pixels to get a photo realistic picture. I was going to start from known photo and work up but as Junior Chubb said it was of the realm of the movies. Then Billpacer explained how it could actually be done and I now I can understand why it's only something that could done the movies. Maybe this could be done in the future with organic or quantum computers. It blows my mind that any photo that can be imagined could be created by manipulating pixels.

1 person likes this

Share this post


Link to post
Share on other sites

That is what I was trying to get at. I wanted to some how manipulate pixels to get a photo realistic picture. I was going to start from known photo and work up but as Junior Chubb said it was of the realm of the movies. Then Billpacer explained how it could actually be done and I now I can understand why it's only something that could done the movies. Maybe this could be done in the future with organic or quantum computers. It blows my mind that any photo that can be imagined could be created by manipulating pixels.

The amount of programming and processing power required for that would be huge. I can imagine how it could be done but it would be very, very hard. You would need a big team of experienced programmers for a job like this. It would also involve scanning lots of images to assist in the prediction software. Of course things will always be made more efficient but we'd be looking a quite a resource extensive program.

I imagine the output would be an array of images with the different predicatibility methods applied to it based on any implemented data like colour set, sizes, etc. Where the user has the option of selecting the one that appears like the one they envisioned or an option to regenerate another set of images.

Share this post


Link to post
Share on other sites

"painting with windows Paint" use that for a Google search to see what you can do then try other search params, its amazing what you can achieve with such a basic software....

Share this post


Link to post
Share on other sites

Assuming you a have workspace that doesn't go on for infinity, there will be set number of pixels. Since each pixel can only be assigned a limited range of colors there is finite number of of combinations that can occur but it will be a huge number. Suppose you have a workspace of only 1 pixel, with 256 different colors that equals 256 different combinations. A workspace of 2 pixels would be 2^256 combinations. A high definition photo that is 1280 x 720 pixels would have 921600^256 different combinations. It is only a matter of being able to get every possible combination. If you were able to do that no only would you be able to do the things you asked but you would have essentially any photo that could ever exist.

Let me make sure I have this correct. It is somehow possible to get an actual photograph from a person, place or thing that was around well before cameras were invented?

Share this post


Link to post
Share on other sites

It would be possible to render any photograph. However, it would be just an estimate unless you had a clear idea how someone or something actually appeared. I could show some random photo of a woman and claim it was Joan of Arc. The main issue is the number of renditions, huge number.

Share this post


Link to post
Share on other sites

Since every photo is composed of pixels. I wonder how difficult it would be to either take a low definition photo and to fill in the gaps to make it high definition or take known photo using various set points to reproduce the photo from various angles. I tried to use a random number generator to assign various colors but it turned out to be a mess. The only other idea I had is use some type of fractal program but my programming skills are not that great. Any ideas?

Try Perfect Resize, not cheap but comes with a 30 day free trial. Up to 3x4 times original size it is very useful.

What you cannot expect is that portions not reflected on the low res imkage suddenly appear by magic on the high res. You will never get more than what existed to start with.

1 person likes this

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0

  • Recently Browsing   0 members

    No registered users viewing this page.