Sora ヽ(´ー｀)ノ

back in teh 2023 on teh geyuriz there was a thread about a niche *chan-analogue that was about islam, does anynigra remember the name?

Heyuri-tan has come to life and exists in Tomodachi Life now

I'm going to post a big update to my vocal synthesis project tomorrow. Stay tuned.

Hint: This image has been driving insane for the last few days.

lets go to this japanese restaurant i found man

why do I have to wait so so soo many months to change my username on vidlii?

Its not fair kaguya

Long ago i was driving on a busy street and on my lane there was a cat seizing up after being hit by a car. All the cars were avoiding it and so did i.
Would you also avoid this cat? Or would you run him over to shorten his suffering but risk outrage from all the onlookers?

Masterpiece

（　´∀｀）つ yuuyuuu

uploading a random video as a new thread since it keeps giving me error as im trying to make a thread with a mp4 file under 10 mb. no idea what the issue may be (i tried recoding the video and ditterent devices) so i wanna see if the issue persists with other videos..admins are free to delete this thread but you can also discuss under it if experiencing similar issues （ ´,_ゝ`）

i've nothing to post or say, still, haev a chen day!

i see...
chibi human is doing something so terrible that the twin dolls are becoming hostile

https://github.com/victorvde/jpeg2png

There's this rather obscure tool on github from a decade ago called "jpeg2png" that's designed to remove jpeg artifacts. I've been using it for a decade now and nothing else even comes close to it in terms of quality. Even those expensive neural network based ones are considerably less effective than this tool.

The only real flaw of it is that some images that actually do contain significant noise and are saved at low quality get smoothed out slightly. Understand though that importantly, that this is not smoothing in the sense of blurring, but reduction of the JPEG coefficients. In general, this tool does not result in any blurring. I believe the tool works by tweaking the DCT coefficients to result in the minimum noise and maximum coherence, which usually results in a result very close to the original.

The reason the smoothing happens for high noise images is presumably because it is because the solution it finds is just to tweak every DCT coefficient so it is slightly smaller to minimize the cost. I wonder if this issue could be simply fixed by detecting when most DCT coefficients decrease, and not changing them in that case. In cases where they are legitimately being decreased, the reduction in cost would be significantly higher and the changes in DCT coefficient should be pretty much evenly spread, as the errors were introduced by quantization/rounding, which is not biased.