Hacker Read

koalaman · 2012-07-31 14:07:24+00:00

couldn't you just encode the images into text?

untog | karma 43841 | avg karma 4.88 · | 2015-06-07 22:19:33+00:00

Huh. Surely there's a better way to do that than post text in an image? Even just using pixels to encode the text.

drblast | karma 3624 | avg karma 6.19 · | 2014-09-19 09:40:13

If only there were some automated way to convert those images to text.

LarryMade | karma 64 | avg karma 0.72 · | 2012-02-01 04:55:21+00:00

Many languages support creating image maps and then writing strings to text to it. Not that hard a process.

artninja1988 | karma 825 | avg karma 2.35 · | 2023-10-13 15:18:13

Would it be impossible to train their own text encoder on just the images they have? How many would one need?

Mirioron | karma 4923 | avg karma 2.01 · | 2020-03-06 23:21:59+00:00

Or just regular text. Or algorithms with PNG or binary blob seeds.

yccs27 | karma 1021 | avg karma 3.08 · | 2021-04-22 10:21:54+00:00

you simply have a tool which extracts the text, transforms it however you please, and then puts it back as an image.

Then you still need some way to handle text, and image to text is not reliable. Text simply has so much more distilled information. Images are nice for humans, but I can't imagine them as a storage format for programs.

reply

Jack0614 | karma 29 | avg karma 1.71 · | 2017-01-30 16:38:00+00:00

Yes, you can make text become image

tomjen3 | karma 9638 | avg karma 1.0 · | 2019-07-24 14:17:38+00:00

How do they manage to do that? Embed the text as an image?

hgsgm | karma 1743 | avg karma 0.72 · | 2023-03-11 09:11:20

Why would you say that? Every word can be easily encoded in a tiny picture.

theGnuMe | karma 918 | avg karma 0.67 · | 2023-02-19 08:31:07

You can go from an image to text and vice versa. People have done it.

kijin | karma 10374 | avg karma 2.8 · | 2013-06-30 10:43:19

Image extraction would be cool, but to me getting a readable block of text is more important.

flippyhead | karma 2704 | avg karma 3.79 · | 2013-04-27 14:05:35

Yes! I love the idea of being able to find images that actually include the text you want. Even better if the image itself is relavent.

londons_explore | karma 35497 | avg karma 2.72 · | 2021-04-22 10:08:06+00:00

Is it time to admit that simply storing everything as an image is the future?

Sure, it isn't technically as neat, but it gives far more flexibility. It happens already in memes (where text is usually part of a gif)

Text and text encoding dates back from the days when every byte of storage space mattered. Now lossless high resolution images are the norm, and wouldn't constrain what symbols could be included. If you want to make a word be slightly bendy for style, you can!

Accessibility, relayout, etc. can all be done with images too - you simply have a tool which extracts the text, transforms it however you please, and then puts it back as an image.

reply

mattway | karma 6 | avg karma 0.75 · | 2021-08-05 21:04:41

The text to image results look pretty awesome, but I feel like they all have a "neural net" quality to them. Whilst the style transer results come out more unique. Do you think it would become possible to combine the two techniques? Eg. Generate me a pic of this text, but in this images style?

teruakohatu | karma 8079 | avg karma 4.97 · | 2023-03-14 13:30:44

My understanding is thta the image embedding is included, rather than converting to text.

SandB0x | karma 11954 | avg karma 12.77 · | 2015-04-25 17:34:18

If only there were some way to describe in a few lines of text what an image should contain?

y04nn | karma 979 | avg karma 2.93 · | 2022-05-23 16:33:33

Really impressive. If we are able to generate such detailed images, is there anything similar for text to music? I would I though that it would be simpler to achieve than text to image.

tsiki | karma 151 | avg karma 2.56 · | 2012-07-31 09:19:47

That's true. But, it's also completely possible to make a machine learning algorithms to separate "real text" (text meant for human to human communication) from text encoded images since they differ in significant ways. Granted, you could always try to make a text encoding format which resembles real text, but I'm fairly sure the machine learning algorithm could be constructed in such a way to make its usage unfeasible.

dash2 | karma 7019 | avg karma 3.37 · | 2022-07-21 07:57:38

How do you generate the original image? And what about the subsequent images, do they come automatically from the text? I'd love to know more about the process.