Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Project Naptha: a browser extension that enables text selection on any image

projectnaptha.com

Post reply on HN
Print view View original

11–20 of 141 posts

Re: Project Naptha: a browser extension that enables text selection on any image

#12

Post by antimatter15 »

3JPLW wrote: post #10

The biggest thing I'd like to see is enabling in-page (control/command-f) search. In my quick scan through the page it looks like it doesn't do that… is that right? Are there plans to add invisible text to the DOM that control-f can find?

One problem with that is that it processes images lazily. It continually extrapolates cursor moments ~1 second into the future and processes those relevant parts of relevant images. But it should be possible that after an image is processed (or even eagerly by looking up previously recognized regions from the cached OCR server), the page could be made Ctrl+F-able.

Re: Project Naptha: a browser extension that enables text selection on any image

#14

Post by antimatter15 »

aalpbalkan wrote: post #11

Certainly a cool idea but it didn't work fine on an XKCD comic: http://www.xkcd.com/ bottom line here is recognized as: "T1EN°5'lI'ONAl.1?E£ONNH\56PNCE(YHCEPlP6fiN(N)SURLH’PR3AO-i‘lDlsIr'£7E‘5IJ%z"

Randall Munroe's handwriting is a bit difficult to OCR because a lot of the letters are smushed together close enough that the it's not possible to unambiguously segment the text into distinct letters (which is a necessary first step in any OCR engine that I'm aware of). Maybe Google's (or Vicarious's) magical convolutional neural net that can solve CAPTCHAs would fare better.

Re: Project Naptha: a browser extension that enables text selection on any image

#18

Post by mcpherrinm »

antimatter15 wrote: post #4
jgj wrote: post #3

> Unfortunately, your browser is not yet supported, currently only Google Chrome is supported. FF 28 seems to be working fine with the "Weenie Hut Jr." version...is it just the add-on that isn't supported? awesome tech, btw

Yeah, I just haven't gotten around packaging the whole thing as a Firefox Addon. It's actually technically possible to run the whole thing on a normal unprivileged webpage (in fact, that's my development environment).

If all you need to do is inject Javascript into a webpage, you should be able to make a Firefox addon to do that in easily under an hour -- check out the SDK getting started guide at https://developer.mozilla.org/en-US/Add-ons/SDK/Tutorials/Ge...

Re: Project Naptha: a browser extension that enables text selection on any image

#19

Post by antimatter15 »

iooi wrote: post #16

Couldn't get it to work on: http://graphics8.nytimes.com/adx/images/ADS/37/09/ad.370964/... Also for: http://www.wsoddata.com/clients/8bec9b10/ads/300x250_static/... It can't get the top-right text correctly Awesome tech though

One of the rules for the heuristic for what images to ignore is that it needs to have over 19,000 square px, and that first image was a bit under that.
Post reply on HN