Automatically “block” people in images using a pretrained neural network

The road to hell is paved with good intentions. And I think software engineers are this century's key perpetrators of "good intentions." Here's a hint: if you saw it in a dystopic thriller you probably don't need to build it, even out of morbid curiosity.

This would combine well with an in-painting algorithm-- or the video equivalents: <a href="https:&#x2F;&#x2F;research.adobe.com&#x2F;project&#x2F;content-aware-fill&#x2F;" rel="nofollow">https:&#x2F;&#x2F;research.adobe.com&#x2F;project&#x2F;content-aware-fill&#x2F;</a> <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=j3uCV0JYMJ4" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=j3uCV0JYMJ4</a>

Why are we building the technology of a dystopia? What happened to shows like Star Trek which gave us more helpful goals to work toward?

“I thought what I&#x27;d do was, I&#x27;d pretend I was one of those deaf-mutes.”

This doesn&#x27;t seem to work for me. I ran &quot;python person_blocker.py -i images&#x2F;img1.jpg -l&quot; after installing dependencies and I only get:<p>&#x2F;home&#x2F;d33tah&#x2F;virtualenv-py3&#x2F;lib&#x2F;python3.6&#x2F;site-packages&#x2F;h5py&#x2F;__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.<p><pre><code> from ._conv import register_converters as _register_converters </code></pre> Using TensorFlow backend.<p>2018-03-31 12:53:18.777911: I tensorflow&#x2F;core&#x2F;platform&#x2F;cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA &#x2F;usr&#x2F;lib&#x2F;python3&#x2F;dist-packages&#x2F;scipy&#x2F;misc&#x2F;pilutil.py:480: FutureWarning: Conversion of the second argument of issubdtype from `int` to `np.signedinteger` is deprecated. In future, it will be treated as `np.int64 == np.dtype(int).type`.<p><pre><code> if issubdtype(ts, int): </code></pre> &#x2F;usr&#x2F;lib&#x2F;python3&#x2F;dist-packages&#x2F;scipy&#x2F;misc&#x2F;pilutil.py:483: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.<p><pre><code> elif issubdtype(type(size), float):</code></pre>

What&#x27;s next, cameras that refuse to take pictures of specific things? I find this unsettlingly creepy.

This reminds me of the character <i>Laughing Man</i> in the anime Ghost in the Shell:<p><a href="https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Ghost_in_the_Shell:_Stand_Alone_Complex" rel="nofollow">https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Ghost_in_the_Shell:_Stand_Al...</a>

In some cases the shape is too small and left a few pixels around the person, or it has corners that are too sharp. Is it possible to add an option to enlarge the shape a few pixels?<p>[Bonus points for another option to make soft shadows near the border of the shape, so it is not so sharp.]

We are now a few steps away from Peril Sensitive Sunglasses. [1]<p>1: <a href="http:&#x2F;&#x2F;www.hhgproject.org&#x2F;entries&#x2F;perilsensitivesunglasses.html" rel="nofollow">http:&#x2F;&#x2F;www.hhgproject.org&#x2F;entries&#x2F;perilsensitivesunglasses.h...</a>

What purpose does this serve? I&#x27;m genuinely curious what are the applications of this?

Prepare your bicycle!!!