Hacker News new | past | comments | ask | show | jobs | submit | zengineer's comments login

congrats - very interesting and well done platform! :)


Would love to use it for my startup, but I believe it is to self-host on a server with GPU? Or is there an easy to use API?


I ran it with 3040x3040px images on my MacBook M1 Pro in about 9 seconds + 200ms or so for the masking.


Previous SAM v1 you can use e.g. in here:

https://fal.ai/models

https://replicate.com/

You just have to wait probably few weeks for the SAM v2 to be available. Hugging Face might also have some offering


It's OSS, so there isn't an "official" hosted version, but someone probably is gonna offer it soon.


That's sweet! Like the UI and threads approach.

How does it generate 3d models?

Btw there is also https://frvr.ai for more casual games


Thanks! We're using a service called Meshy (https://www.meshy.ai) to generate 3d models.


I realized that I never read anything on Medium. I use it solely as my private travel blog, which is then read by family members, close friends and my nostalgic future me. The reason I like to use medium for that is that it is very easy to add content. The two biggest issues I have with Medium is that pictures can not be set fullsize anymore and that there is a (for me) useless sidebar. Not only are you now limited on how to present your content, but it adds a lot of distraction.


Very cool and great start!

I tried it and I wonder how I could generate more images. I made a story and only had one image.

I would also love to print it out as a book (don't see myself reading a book to a kid on a laptop / tablet) and that every page would have an image. Would also be good if I could define which content should be represented as an image.

Keep up the great work!


Hey!

free stories only offers 1 image, and paid accounts you'll get 5 images per story. We are working on full Edit an upload images capabilites. https://twitter.com/LinusEkenstam/status/1617997909508902913

Paid accounts can also re-generate images using their own prompts for images.

So we are working on distribution, printed books, e-books, audio and video. It's all coming in the next few weeks.

Thank you for the kind comment!


I am highly interested in the Frontend position, but it doesn't seem to be listed on the website. Thank you!


Wow, this is how a digital trip to a museum should be! Really well done and interesting - not giving too much information at the same time.


I hope they add a "remote working approved" tag and filter, just like the "superhost".

I have been working remotely since almost a year now in several European countries and to filter out AirBnBs, which offer a decent desk and have stable internet takes hours. The "workstation" filter can not be trusted and neither the "has Wifi", which doesn't say anything about the quality.

An internet speed test should be required to be done by the host. This way I can avoid having to ask about the internet every time before booking - which takes sometimes half a day for getting a response.


I don't think I'd ever book a place that didn't explicitly list Internet speed if I were to work there and I hate that AirBnB doesn't easily provide this information. The last time I booked a place like that they claimed to have "professional grade" Internet which ended up being a 15mbps down/ 4mbps up connection. Surprisingly (to me at least) that was enough for most video calls.


The problem is that in a lot of Western Europe, an actually-good Internet connection is a very rare commodity - DSL is still a thing in 2022.


There are a few countries where broadband is unjustifiably bad (Germany and Belgium come to mind) but I wouldn't generalize to the entire Western Europe. For instance many rural departments of France have extensive FTTH coverage.

Also DSL isn't necessarily so bad, G.fast allows for speeds that are close to what you can do with modern Docsis deployments.


This article actually "saved my life". I read it back then in 2013 and in 2014 I almost drowned.

Just before I felt that something was very wrong I remembered that this post said that if you drown it is already too late to scream for help. So I shout to my friend, who was also with me in the water, that I am about to drown. He then tried to help me but then he struggled too because of the current and the strong waves which came out of nowhere. It took us a lot of effort to get back to a point where we were able to stand. Until today I try to avoid to go deeper than I can actually stand.


For your last sentence, do you mean "to this day"? "Until today" means something that ended/stopped today. Or are you a good enough swimmer now that you feel comfortable swimming in deep water?


Yes, to this day.

Thanks! :)


Is there platform like StackOverflow to ask more about best practices / architectural issues?

The other day I gave a very detailed question on StackOverflow about solving a specific problem in computer vision. I have little experience in CV, but wanted to know from experts into what I should look into or how they would tackle this problem / find other edge cases I were missing etc. My question got deleted since I wasn't able to provide any code, but more of a "This is how I imagine it to do" and that was not a coding question. I tried reddit, but there was no real activity.


Yes, actually from the same Stack Exchange family, you can try with: https://softwareengineering.stackexchange.com


Thanks! Completely forgot that this was there too


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: