Anne van Kesteren

Google reads images

Google Catalogs Search: xhtml css. Nice!


  1. Wow, that's just brilliant. Will the complete text of the images be available as text as well?

    Posted by Asbjørn Ulsberg at

  2. That's the right comment: "Wow!" That algorithm is not that trivial, I think.

    Well done, Google!

    Posted by Christian Machmeier at

  3. I had noticed this just a couple days ago: a search for a technical subject brought up a PDF in the results, but this PDF was a scan, not a text PDF. Pretty amazing. You can even use Google's "view as HTML" option, though it's very hard to read.

    Posted by Adam Rice at

  4. holy crap that's awesome.

    Posted by Jeff Minard at

  5. I wouldn't want to spoil your enthusiasm but this has been possible with Google for ages, and even they were late to the party. So I don't really see what the fuss is about.

    Posted by Frank Geerlings at

  6. [...] Anne hat die Blogosphäre darauf aufmerksam gemacht, dass Google jetzt Bilder lesen [...]

    Posted by DenkZEIT :: Google kann Bilder lesen at