Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Riskiness score underestimates visible distortion #14

Open
sp-1234 opened this issue Dec 4, 2017 · 1 comment
Open

Riskiness score underestimates visible distortion #14

sp-1234 opened this issue Dec 4, 2017 · 1 comment

Comments

@sp-1234
Copy link

sp-1234 commented Dec 4, 2017

sjpeg recommends sharp YUV for this image (https://user-images.githubusercontent.com/19407943/33548263-53136816-d8ef-11e7-8994-6a6cfded7912.png):

Riskiness: 60.1 (recommended yuv_mode: SharpYUV420)

But in fact only YUV444 gives transparent results for this image; any chroma subsampling adds very obvious changes to the image.

@skal65535
Copy link
Collaborator

Yes, this image is very challenging.

The threshold between SharpYUV420 and YUV444 is set at the riskiness score of 70 (see kThreshSharpYU420 in jpeg_tools.cc:186).. Ideally, the threshold should be made dependent of the targeted quality. The current setting corresponds to some sweet-spot around quality 70-80, which is obviously not enough for this particular difficult image.

hope it helps,
skal/

Note that 'vjpeg' can show the riskiness-map (instead of just having the total riskiness score) by pressing the 'r' key. Here's the result:

vjpeg_risk_map

As can be seen, all of the image is problematic. Usually only few areas are (crisps edge, etc.).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants