SmartTextExtraction looks like exactly what I need! However, I have to be able to run offline.<p>How can I download it and run the algorithm locally?<p>Tesseract gave me pretty poor results, probably because I'm using photos of printed pages instead of flatbed scans.
Finally registered to Hacker News just to say thanks for this. I wanted to use Google Cloud for so long but because I'm in Europe, I can't use it for non-business purpose. The algorithms I'm interested in seem to work pretty well so far for what I tested (text detection).
<p><pre><code> https://news.ycombinator.com/item?id=15965330
</code></pre>
This was featured on HN a few days ago, it uses artifacts from the manufacturing process to extract data about which device took a photo. Is there any sort of algorithm on algorithmia at the moment that can give you this kind of information? Thank you, loved the post.
There is some overlap with the IPOL journal : <a href="http://www.ipol.im/" rel="nofollow">http://www.ipol.im/</a><p>It is an image processing journal where the source code of the articles is peer-reviewed, and all the algorithms can be tested online.