Automatic Image Colorization with Simultaneous Classification

For context, a different colorization model with about the same results: <a href="http:&#x2F;&#x2F;richzhang.github.io&#x2F;colorization&#x2F;" rel="nofollow">http:&#x2F;&#x2F;richzhang.github.io&#x2F;colorization&#x2F;</a><p>Another model previously posted on HN, with (IMO) worse results than these two models: <a href="http:&#x2F;&#x2F;tinyclouds.org&#x2F;colorize&#x2F;" rel="nofollow">http:&#x2F;&#x2F;tinyclouds.org&#x2F;colorize&#x2F;</a>

This is amazing to me. My major was Digital Imaging Technology in 2005 and I remember doing this by hand in photoshop wondering if one day there would be a button for it.

Tried with some historical photos and my own BW images. It&#x27;s missing the global image prior for most of the images except vegetation, it has a hard time even with people. For local features I&#x27;ve seen similar problems, my guess is that they trained it on a not large enough dataset, the spectacular samples are from over-training. While the idea looks promising the current implementation is far from general.

A small sampling of how this preforms on some B&amp;W images I had laying around (all my family)....<p>My Grandfather &amp; Brothers: <a href="http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_0090.jpg" rel="nofollow">http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_0090.jpg</a> <a href="http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_0090.color.jpg" rel="nofollow">http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_0090.color.jpg</a><p>My Grandfather, My mother and my Aunt <a href="http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_4629.jpg" rel="nofollow">http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_4629.jpg</a> <a href="http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_4629.color.jpg" rel="nofollow">http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_4629.color.jpg</a><p>My grandfather and my grandmother: <a href="http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_6868.jpg" rel="nofollow">http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_6868.jpg</a> <a href="http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_6868.color.jpg" rel="nofollow">http:&#x2F;&#x2F;adam.gs&#x2F;v&#x2F;IMG_6868.color.jpg</a><p>From my perspective, these are decent results considering what they have to work with, I think it did a very good job.

Impressive stuff, I especially like the style transfer that can be done by using the global features of one image and the local features of another (Fig. 7)<p>What I find somewhat annoying is that whilst they show some examples from their validation set, and a couple of examples of the model failures. They don&#x27;t appear to show a <i>random</i> selection of cases from their validation set.

I&#x27;m surprised by how apt Lua is for these kind of algorithms. From the architecture diagram I expected to be hit by a large blob of code but found that most of things are taken care by the language&#x2F;framework itself!

Can someone please put the file colornet.t7 on the torrent network or a high-volume service somewhere? I&#x27;m probably not the only having a hard time downloading that file.

I would really like to see approaches like these applied to movie scenes. Especially how differences in single colorized frames depicting the same scene could be handled.

I wonder how companies doing this for money will be impacted. That&#x27;s amazing what a time saver it will be.

Not sure if the title would be too long, but I&#x27;ll be honest and say that I thought this was about the news organization for a minute.<p>CNN = Convolutional Neural Networks in this context.