This repository has been archived by the owner on Dec 7, 2023. It is now read-only.
Weber default algo #264
Closed
Weber default algo #264
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…enerated means, variances, and covariances
…eber'
BREAKING CHANGE: If users depend on {ssim: 'fast'} as the default, their test cases with this
library will fail
omnisip
force-pushed
the
weber-default-algo
branch
from
5dc0a99 to
bb073bc
Compare
…ize mean and variance Fixed a few calculation issues. Still have an outstanding issue with respect to isabe. For some reason this particular image iskewing the SSIM results.
…per rounding of the input Current input luminance input is not rounded prior to the ssim function. This means that we can't use an integer implementation and expect the results to work properly until the numbers are normalized. As such, I've implemented a Uint8Array wrapper that rounds all of the luminance values passed in.
…egers Uses Uint32Array instead of Uint8Array for rounding during the SSIM calculation
Removed unnecessary debugger line. Added comment to show how variance tolerances were calculated.
… src/weberSsim.ts Fixed LICENSE to be more accurate of contributors
… src/weberSsim.ts Fixed LICENSE to be more accurate of contributors
…st one weber ssim
Separated the main project license file from the license files for the contributions
|
|
|
Weber is now the default 🎉 as part of #273 |
|
Score. It's just in time for performance improved version too. :-)
If the JS works like my new scalar and simd version, it should shave 50% or
more on times for partial sum calculations. In C without vectorization, it
partial sums an entire 1980x1080 frame in 6.9ms (no parallelism) and 5.8 ms
with simd (also no parallelism). The old algorithm scalar implementation is
13.9 or 14ms.
I redesigned the sum algorithm to be dually parallel and have less than 1/4
operations. It's also more space efficient (only requiring WxH memory per
sum buffer instead of W+1xH+1.
…On Wed, Sep 2, 2020, 06:13 Oscar ***@***.***> wrote:
Weber is now the default 🎉 as part of #273
<#273>
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#264 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACKQJXKFLFVH26XS5K52SLSDYZFFANCNFSM4P7BUSNA>
.
|
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Since this is a breaking change I broke it into a separate pull request.