Icons8

5 ms·

1.
My $120 GE smart scale's body fat is just 1.5 × BMI − 17.5(aiandtractors.com)
8 points by Icons8 21d ago|0 comments
2.
by Icons8 3mo ago
That was a satisfying part of my day. Thank you.
3.
by Icons8 3mo ago
Most passport rankings count every country as 1 point, so Tuvalu = France. I made a version where each visa-free door is weighted by how many people actually go there. So opening France, Spain, the US, China, Turkey etc matters much more th
4.
Show HN: Passport rankings weighted by where people travel(aiandtractors.com)
3 points by Icons8 3mo ago|1 comments
5.
by Icons8 7mo ago
Health data. That’s why they’re rolling out Chatgpt Health.
6.
by Icons8 8mo ago
https://aiandtractors.com/ - the last iteration of my personal website (the first one was in 1996). I mostly use it to vibe code the hobby projects; the last one is a better passport ranking that brings Korea to the first p
7.
AI fallout: businesses erased people from their visuals(aiandtractors.com)
3 points by Icons8 1y ago|0 comments
8.
by Icons8 2y ago
Median is $360k https://www.salary.com/tools/salary-calculator/ceo-non-profi...
9.
by Icons8 3y ago
There's British though
10.
by Icons8 3y ago
Over self-censoring
11.
by Icons8 3y ago
Stock photography, better selfies, or simply fun. Also, people always invent some use creators never thought about.
12.
by Icons8 3y ago
Train their models
13.
by Icons8 3y ago
We just released our guide for programmers looking to enhance their programming efficiency and productivity using the power of ChatGPT. What’s inside: - Comprehensive guide with practical tasks and ChatGPT coding examples - Demo ChatGPT con
14.
by Icons8 4y ago
We've built a very similar feature across https://icons8.com content. What keeps exciting me when you upload a photo of a building and find tiny icons with a somewhat similar architecture, or illustrations, or realistic ren
15.
by Icons8 4y ago
Founder here. You have correct observations and ask the right questions. 1. We do promote our content. It's not enough to support the development. 2. We could afford a generous freemium. Unlike Figma, Lunacy is resource-efficient and h
16.
by Icons8 4y ago
Their specs help, but we also support their old format which was basically a dump. We had to reverse engineer it. Also, we'll import from Figma this fall. You're right, it's never easy. It's a bug valley of countless fil
17.
by Icons8 4y ago
Founder here. Could you elaborate on rough typography, please?
18.
by Icons8 4y ago
Founder here. A little bit, yeah. Twice the usual installs in the Microsoft Store: https://lun-us.icons8.com/a/mI9C5AKookGuZXtrXLg1Sg/HNVJ3hin3... Thank you for supporting us; I'm touched.
19.
by Icons8 4y ago
Thanks for mentioning.
20.
by Icons8 6y ago
I see your concern and it’s valid. We don’t store anything, but you probably need more warranties than that. I don’t think it’s the best way to hide. Generated photos are detectable after all, and using a generated photo raises a flag. But
21.
by Icons8 6y ago
It's not. You can use it for free, just attribute it. $2.99 gives you a double resolution. We don't use these photos to train.
22.
by Icons8 6y ago
I’d like to see the source photo if you don’t mind. I’ve discovered that in any unclear situation it shows toddlers. In particular, it happens when the camera distorts the perspective, such as too close or shot with an unusual angle. Precis
23.
by Icons8 6y ago
Right! Or Google Maps — enough of people with the heads in a cloud.
24.
by Icons8 6y ago
Some of the questions: How is it done? The images are pregenerated: 2.6M to be precise, all of them are here: https://generated.photos/faces Technical challenges of better resemblance We could either pre-generate (see t
25.
by Icons8 6y ago
Founder here. Thanks for featuring! There was an improvised AMA in another thread: https://news.ycombinator.com/item?id=25208704 I'd be glad to answer the questions here as well.
26.
by Icons8 6y ago
Haha I like how you put it! On the boring side, it’s probably not worth it. Generated photography is detectable. A trained eye could spot it; big social media have models trained to a lot it. Scraping a real life dataset makes more sense t
27.
by Icons8 6y ago
Thank you for buying and thank you for your encouragement! My personal fault was positioning. We can’t make everyone looking similar. As you said, we have a bunch of files on a disk, heavy indexed ones. We spend a year working on diversity,
28.
by Icons8 6y ago
I like the idea! That would be a cool next step after this mvp! We could either pre-generate (see the first comment - it’s an accurate guess), or generate just in time. The first is way easier - same architecture, more storage (we are movin
29.
by Icons8 6y ago
Hi, founder here. You're right about the pre-generated images! 2.6M to be precise, all of them are here: https://generated.photos/faces It definitely has some fitting issues; Brad Pitt is a good example as he dyes his
30.
by Icons8 10y ago
Great fork, thank you!

More ›