Jump to content

Is it possible to resample 1080p 4:2:0 8bit in something better in 720p?


thementalist
 Share

Recommended Posts

Come on, you had to expect this next logical question, after the 4k to 1080p resample.

I know, 1080p is not four size of 720p. So let's start theory for poor 1080p-people like me who doesn't have money to get 4k.

Today, DSLR shoot in 1080p in 60 fps if we are lucky, but it's enough or could it be. To get better colours, can we use the trick to resample in 720p, in a certain way?

If it's a stupid question, sorry

Link to comment
Share on other sites

EOSHD Pro Color 5 for Sony cameras EOSHD Z LOG for Nikon CamerasEOSHD C-LOG and Film Profiles for All Canon DSLRs

_If_ 4K 4:2:0 really turns into 1080p 4:4:4, then 1080p 4:2:0 makes 540p 4:4:4. So downscaling to 720p would result in something between 4:2:0 and 4:4:4 (4:2:2 I guess?).

 

I don't really see why you'd want to do so though, unless you are watching on a 720p screen it will be upscaled again to 1080p, introducing artefacts and degrading the image quality.

Link to comment
Share on other sites

When you go from 4K to 2K, you have a 2:1 ratio in each direction (vertical and horizontal).  When you go from 1080p to 720p, you have a 3:2 ratio.  You don't quite have enough pixels to have a proper 4:4:4 subsampling, but you could use some sort of averaging or binning of pixels to interpolate a greater amount of chroma detail.

Link to comment
Share on other sites

Can the improvement justify this discussion and the 'how to'?

Thanks

 

Here's the gist of how it would work.  Let's take a quick look at what a 4:2:0 file looks like:

 

NvtYHl5.png

 

The black lines represent pixel borders.  Here is that same image broken into luma and chroma components:

 

FGtQzf3.png    hfl81Ny.png

 

As you can see, the luma channel has data for each individual pixel, whereas the chroma channels only have data for groups of 4 pixels.  In other words, the chroma channels are at a lower resolution (half the resolution vertically and half the resolution horizontally).

 

Now, if we scale down the image, we should get something like this:

 

SMheIoQ.png

 

(Instead of scaling the image down, I just resized the pixels)  Remember, the black lines denote the pixel borders.

 

So we end up with the original values spanning several pixels.  Obviously each pixel can't contain more than one colour value, so we have to take the weighted average of each pixel.  We end up with this for the chroma channels:

 

O7uLUXn.png

 

As you can see, we're getting a different colour value for each pixel.  This technique isn't as accurate as downscaling 4K to 2K, but it does give you a pseudo-4:4:4 image.

 

The luma channel would also be affected.  But this far into my response, I realize that the luma pattern I used in my examples was a bad choice and doesn't properly illustrate what happens.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • EOSHD Pro Color 5 for All Sony cameras
    EOSHD C-LOG and Film Profiles for All Canon DSLRs
    EOSHD Dynamic Range Enhancer for H.264/H.265
×
×
  • Create New...