Use google reverse image search to find a high res version of a pic

>use google reverse image search to find a high res version of a pic
>save it
>.webp
>delete it
>never use that shit again

Attached: 1542640647857.png (970x545, 408K)

Other urls found in this thread:

s000.tinyupload.com/index.php?file_id=80686111421098077344
twitter.com/NSFWRedditVideo

Can't stop the future faggot.

Google image reverse search sucks the gayest balls

Tin eye is alright tho

convert image.webp image.jpg
wow sure was difficult

tin eye has let me down too many times in the past, so I haven't touched it in years. I do use yandex when I don't wanna bother with google.

>recompress a webp conversion of a lossy image format
do you hate data?

Yes

Attached: 1537783853799.jpg (499x499, 6K)

>upload 2.1 KB file
>Jow Forums makes it 6 KB
kek what

gook has to inject his botnet into everything

>right click
>save as
>jpg

it's still a webp you mongoloid. meaning you can't open it in software that does not read webp

Apparently Jow Forums has something to classify images?

Attached: 1531764626754.png (705x175, 38K)

download the Jow Forums version of your file, and compare it with the original 2.1KB version

It's from the Jow Forums version, it wasn't present in the original.
s000.tinyupload.com/index.php?file_id=80686111421098077344

this deserves its own thread so we can find out wtf is going on with Jow Forums/gook injecting shit in our files

>software that does not read webp
So useless boomer tech

Attached: 1548711555054.jpg (906x1024, 66K)

Attached: 1552697809517s.jpg (110x125, 3K)

They're interjecting a moment to say, that Linux is not GNU/Linux, but rather a...

Jow Forums recompresses JPEGs losslessly to strip any EXIF data.

>recompresses JPEGs losslessly
>recompresses
>jpeg
>lossless

Okay kiddo

But how do you explain ?

test

Attached: 123.png (184x179, 9K)

Welp, thought I deleted the losslessly part before posting. Guess, I should go to bed.

its nothing wory about guys just ignore

why is this so funny?

You're fucking stupid, that webp image has HIGHER quality than a JPG one.

>this google developed botnet is better, trust me
cringed

tineye is great for finding the biggest image possible if it finds it, but its recognition is kinda ass

>converting a jpeg to webp increases quality
retard

>google image search

Attached: Screenshot_2019-03-09 european people history 10 тыс изображений найдено в Я� (1668x918, 750K)

THE TVRK BULL OWNS GOOGLE. WH*TOIDS ARE KNEELING TO US ALREADY

>open source
>botnet
k

Web servers convert raw photographs into lossy formats you dipshit. When it makes a JPG and Webp file of the same file size the webp image has significantly higher quality. Even at half the file size of JPG, Webp images can look better if converted from lossless sources.

>Web servers convert raw photographs into lossy
are you retarded? google images is fetching the jpegs and converting them to webp. the server doesn't have the raws you mongoloid

No, it isn't. It's simply grabbing the webp image files from the server. Go ahead and look at them, the webp will look better if it's the same file size as the JPG. The only way google would convert said image to webp OTF would be if you used any data saving mode.

why the fuck isn’t there an instagram reverse image search? google/tineye don’t index instagram. I don’t even have an instagram account but it could still come in handy.

it also feels like we should have image searching that isn’t so autistically needy about having to match the same cropping/aspect ratio. sometimes I only have a portion of the image and want to find the whole thing but it never fucking works

Attached: image.jpg (719x719, 267K)

>It's simply grabbing the webp image files from the server.
hey faggot, I can see .jpg in the fucking url. plus, you're implying every fucking site has their images in raw. kys retard

URLs can point to webp images if you have a web browser that supports it, it's up to the server to give you that JPG or not. Spoof user agent to get said JPG and then compare it with the Webp image file. Use emacs pic viewer if on wankblows.

he's trolling you pretty hard

Attached: 3q2oqf.jpg (409x393, 29K)

>emacs
Meant to say numacs

nah, he's a legit brainlet:

PROTIP: you can use the WebM container to house Webp images and post them on 4chinks.

for %%f IN (*.png, *jpg) do (
ffmpeg -loop 1 -i "%%f" -filter:v hqdn3d=8.0:8.0:10.0:10.0 -c:v libvpx -qmin 22 -qmax 22 -quality best -t 2 -r 1 "%%~nf.webm"
)

bat script for windows user, dunno how to make it for loonix users.

test

Attached: 1552695491199.webm (970x545, 13K)

testing with -quality good -cpu-used 0

Some user a while ago said the -g VP8 param does something to help webp in webm. Anybody got anymore info on that?

Attached: 1552695491199.webm (970x545, 15K)

Linux bash script:
mkdir out
for f in *.{png,jpg};
do ffmpeg -loop 1 -i "$f" -g 1 -c:v libvpx -deadline good -cpu-used 0 -qmin 22 -qmax 22 -t 1 -r 1 "out/${f%.*}.webm"; done

Removed stupid filter and 2 second duration. Should work for phoneposters through termux.

wtf, this makes my phone cpu go 100%

>WebP

Attached: webpiss.png (1920x1080, 1.72M)

Wow the CPU is actually doing something, better be worried

Not you again. You can fuck right off with your denoising filter and your inability to understand libvpx's quality control methods.

>Jow Forums still hasn't figured it out
>outsmarted by some gookjew mongoloid
yikes

>10,10 temporal smoothing
>for single images
ah, one of those ffmpeg commands made up of random bits and pieces from other peoples' commands

>use google
you can only blame yourself for being a retard

Does this really have any effect irl where webp would only be transcoded a couple of times at most?

>its literally editing 2 strings in about:config
>this apparently brings the technology board to its knees

Attached: yellow vest boomer.jpg (672x787, 160K)

It's true, I just copy and pasted that from a thread a few months ago. Still pretty nice we don't need gookmoot's permission to post webp images here.

It's inconvinient AF but I'll take it desu.

Attached: To_Aru_Kagaku_no_Railgun_Misaka_Imouto-179976.webm (1440x810, 61K)

Why does -quality good -cpu-used 0 produce higher quality output anyway?

Depends on the image's resolution.
New libwebp versions produce fairly consistent results for low resolution pictures even after being recompressed a thousand times. In fact there's such a big improvement compared to the results in , that it almost seems like the devs saw this comparison and tweaked the encoder accordingly.
For high resolution images lossy WebP starts fucking up colors and smoothing over fine details rather quickly. Although the results never get as bad as with old libwebp versions.

It's only really worth it for pictures that are larger than 4MB and even then you could argue if it wouldn't be wiser to host it somewhere else and link it. Otherwise you just fuck up the quality for saving a few seconds of uploading time.
Also it doesn't replace lossless WebP.

You mean in comparison to -quality best? They produce pretty much the same result, -quality good -cpu-used 0 is just a lot faster as it gives the encoder a generous time limit per frame instead of no limit.

tineye finds mostly exact images it doesn't do any advanced recognition. use yandex for that.

*heats up your cpu*

Yandex is literally black magic

Yandex has been the best one so far. Whenever google has failed, I open up yandex and 99% of the time, it finds exactly what I was looking for.

Not this one, he has t2 r1 which prevents infinite instant looping like does.