NormCap: OCR powered screen-capture tool • Hacker News Board

Re: NormCap: OCR powered screen-capture tool

#21

Post by tough » Fri, Apr 28, 2023, 10:16 PM UTC

Mac only but I am a happy user and can recommend

https://github.com/schappim/macOCR

Just rediscovered the Shortcuts a couple days ago while installing it on a friend's mac.

Re: NormCap: OCR powered screen-capture tool

#22

Post by bjoli » Sun, Apr 30, 2023, 9:22 PM UTC

quelltext wrote: post #15

bjoli wrote: post #2

I had my mind blown by the same functionality in Android. Being able to select text from an app in the app-switching context is amazing. I will definitely use this.

That doesn't use OCR.

It can select text from images. It uses OCR.

Re: NormCap: OCR powered screen-capture tool

#23

Post by sitkack » Mon, May 01, 2023, 4:36 PM UTC

maxlin wrote: post #14

I created a basic similar app for my own use because I wanted to have an idea what people are conversing in russian in online lobbies. WPF, Tesseract OCR and Microsoft's translation API. https://streamable.com/ykng5u A fun side project that I do end up using a bit. Gonna bind the capture to some hotkey so I can use it without changing app focus. Most annoying problem though is that Tesseract OCR often gets confused w…

That is a super neat.

Visual Universal Translator.

Re: NormCap: OCR powered screen-capture tool

#24

Post by holbue » Tue, May 02, 2023, 6:22 AM UTC

Author here, excited to find my tool on hn! Happy to answer any questions.

PS: People looking for (FOSS) alternatives, look here: https://github.com/dynobo/normcap#similar-open-source-tools

Re: NormCap: OCR powered screen-capture tool

#25

Post by holbue » Tue, May 02, 2023, 6:32 AM UTC

csdvrx wrote: post #4

Is there anything on Linux not based on tesseract for the OCR? It's not very good. I miss being able to copy/paste from blurry or deformed screenshots of youtube on Windows.

As others mentioned, Tesseract is SOTA in FOSS OCR. It also still is being developed, improving slow but constantly.

The main issue for a use-case like NormCap are the trained models: they are optimized for images of _printed_ text and layouts, which is different from on-screen-text in many aspects. Unfortunately, I don't have the resources to train my own models.

Cuneiform was a long time competitor, but afaik development there is stalled.