Live data from Hacker News

Building an Internet Scale Meme Search Engine

findthatmeme.com

151–154 of 154 posts

Re: Building an Internet Scale Meme Search Engine

#151

I have absolutely no experience in this area and I'm curious: is there really no open-source text recognition software that's on-par with or close to Apple's (presumably proprietary) implementation? the article mentions Tesseract. is that the current best open-source option?

Certainly not. https://paperswithcode.com/task/optical-character-recognitio...

Re: Building an Internet Scale Meme Search Engine

#152
post #150

If you don't need advanced search features, you can use Sonic ( https://github.com/valeriansaliou/sonic ). It's blazing fast and you can save lot of money on servers.

Someone mentioned lnx.rs in this thread, how does it compare? And how about all the other new Rust solutions like MeiliSearch, Toshi, Quickwit?

I don't know, I am not an expert on this topic. But this could be a nice topic for a blog article that will hit the frontpage on HN :)

Re: Building an Internet Scale Meme Search Engine

#153

Although not as elegant a solution as this I've also tried my hand as well at indexing and categorizing memes. I wanted to save a very specific type of meme though since there are, in my mind, 2 main categories of memes. The first category are what I call "story" memes, they are standalone and typically what you see being shared on Facebook. They usually have text and are able to tell a story on their own with no add…

What are you using as the back end to host your website?
Post reply on HN