Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Real-Time Adaptive Image Compression

wave.one

Post reply on HN
Print view View original

41–50 of 51 posts

Re: Real-Time Adaptive Image Compression

#41

Post by tomaskafka »

trevyn wrote: post #13

Nice work, but disingenuous to not include a BPG (HEVC) image for comparison -- BPG is close to state-of-the-art, not WebP -- even their own SSIM charts show this. Interesting that decoding is slower than encoding. Also curious about performance on CPU. This approach may also be susceptible to "hallucinating" inaccurate detail; you can see a little bit of this on the upper-right of the girl's circled eyelid compared…

> This approach may also be susceptible to "hallucinating" inaccurate detail

Yes! This. I can see us givng up decisions to 'AI' without realizing that it's just a loose association machine. Would I want to have my mortgage rate adjusted by a 'loose association'?

https://twitter.com/keff85/status/862690920805916672

Re: Real-Time Adaptive Image Compression

#42

Post by tomaskafka »

discreditable wrote: post #19

This encoder seems to have some weird distortions that are most visible in the aerial shots. Compare [1] and [2]. The lines on the basketball court are distorted and curved. If you look closely, there is also curvature added to the sidewalks where there isn't any. In case those links break, I'm referring to the top row of aerial images. [1] https://static1.squarespace.com/static/57c8be4459cc68c3e3d7b... [2] https://s…

Yep - and see how it almost disappeared the white car in the shadow in the left middle part. "We've deleted your only alibi so that our data would compress better ..."

Re: Real-Time Adaptive Image Compression

#43

Post by akx »

CyberDildonics wrote: post #25

Does no one realize this is a joke / marketing? Directly from the paper's PDF: "Finally, Pied Piper has recently claimed to employ ML techniques in its Middle-Out algorithm (Judge et al., 2016), although their nature is shrouded in mystery."

Or, you know, that could just be a humorous reference to the TV series, while this is a real implementation.

Re: Real-Time Adaptive Image Compression

#44

Post by dschndstryr »

CyberDildonics wrote: post #30
svantana wrote: post #29

Very impressive work, though it seems like a mistake to focus on compression, which gets less valuable as storage and bandwidth gets cheaper. You need only look to the staying power of jpeg, which is so far from the state of the art, yet it's not going anywhere. Why? The demand for replacing it is not strong enough. They obviously have some good image priors here, if I were them I would consider applying this tech to…

It isn't 'very impressive work', it is marketing for Silicon Valley (impressive marketing though). Literally the first sentence of the linked page: "Even though over 70% of internet traffic today is digital media, the way images and video are represented and transmitted has not evolved much in the past 20 years (apart from Pied Piper's Middle-Out algorithm)." EDIT: This is embarrassing that not one person in this thr…

It's humor. Everything else looks kinda real. The science is bad though, because it lacks all details needed to reproduce and the evaluation is a bit fishy. They also claim it's super fast without mentioning what makes their approach fast.

Re: Real-Time Adaptive Image Compression

#45

Post by vladdanilov »

As someone interested in the field, this does not look much different from state-of-the-art video codecs, e.g. variable-size blocks, wavelet, predicator, arithmetic coding. Only that the predicator is trained on real data. But symbol dictionaries are already used in modern compressors like brotli and zstd.

Most codecs have been tuned for a mean subjective score (MOS). MS-SSIM is not particularly good to fully rely on [1] [2]. In my experiments it performed poorly.

I think Google team effort will have a much bigger impact [3] just by combining all recent practical improvements in the image compression.

Meanwhile, they could have optimized images on the website a little better. Saved ~15% with my soon-to-be-obsolete tool [4].

[1] https://encode.ru/threads/2738-jpegraw?p=52583&viewfull=1#po...

[2] https://medium.com/netflix-techblog/toward-a-practical-perce...

[3] https://encode.ru/threads/2628-Guetzli-a-new-more-psychovisu...

[4] http://getoptimage.com

Re: Real-Time Adaptive Image Compression

#46

Post by web007 »

akx wrote: post #43
CyberDildonics wrote: post #25

Does no one realize this is a joke / marketing? Directly from the paper's PDF: "Finally, Pied Piper has recently claimed to employ ML techniques in its Middle-Out algorithm (Judge et al., 2016), although their nature is shrouded in mystery."

Or, you know, that could just be a humorous reference to the TV series, while this is a real implementation.

This is my read too, since they're citing Judge et al versus the characters or the characters' papers. Its the same vein as when Dropbox's Lepton article cited middle-out as a humorous attempt at self-promotion.

Re: Real-Time Adaptive Image Compression

#47

Post by CyberDildonics »

akx wrote: post #43
CyberDildonics wrote: post #25

Does no one realize this is a joke / marketing? Directly from the paper's PDF: "Finally, Pied Piper has recently claimed to employ ML techniques in its Middle-Out algorithm (Judge et al., 2016), although their nature is shrouded in mystery."

Or, you know, that could just be a humorous reference to the TV series, while this is a real implementation.

So you think they would put a humorous reference to a TV show in the synopsis of their groundbreaking image compression paper and again as an academic reference in their paper?

This whole thread is like being the only sane person in an asylum.

Also, it isn't a 'real implementation' since there isn't any source code to reproduce the results.

Re: Real-Time Adaptive Image Compression

#48

Post by AstralStorm »

tomaskafka wrote: post #42
discreditable wrote: post #19

This encoder seems to have some weird distortions that are most visible in the aerial shots. Compare [1] and [2]. The lines on the basketball court are distorted and curved. If you look closely, there is also curvature added to the sidewalks where there isn't any. In case those links break, I'm referring to the top row of aerial images. [1] https://static1.squarespace.com/static/57c8be4459cc68c3e3d7b... [2] https://s…

Yep - and see how it almost disappeared the white car in the shadow in the left middle part. "We've deleted your only alibi so that our data would compress better ..."

Lossy codec being lossy is a feature, not a drawback.

Re: Real-Time Adaptive Image Compression

#49

Post by AstralStorm »

CyberDildonics wrote: post #28
jstanley wrote: post #27

Earlier quoted context omitted.

I couldn't find that in the PDF?

Are you asking me whether or not you couldn't find it in the PDF? On the actual web page the first line of its abstract: "Even though over 70% of internet traffic today is digital media, the way images and video are represented and transmitted has not evolved much in the past 20 years (apart from Pied Piper's Middle-Out algorithm)." From the PDF: https://arxiv.org/pdf/1705.05823.pdf The end of Section 2.2. ML-based l…

What if the paper was generated using the codec and it is a multilevel joke?

Re: Real-Time Adaptive Image Compression

#50

Post by CyberDildonics »

dschndstryr wrote: post #44
CyberDildonics wrote: post #30

Earlier quoted context omitted.

It isn't 'very impressive work', it is marketing for Silicon Valley (impressive marketing though). Literally the first sentence of the linked page: "Even though over 70% of internet traffic today is digital media, the way images and video are represented and transmitted has not evolved much in the past 20 years (apart from Pied Piper's Middle-Out algorithm)." EDIT: This is embarrassing that not one person in this thr…

It's humor. Everything else looks kinda real. The science is bad though, because it lacks all details needed to reproduce and the evaluation is a bit fishy. They also claim it's super fast without mentioning what makes their approach fast.

No one else seems to realize that.
Post reply on HN