You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When browsing through the posts I don't like to see preview images of human faces, especially politician faces. Human faces elicit automatic emotional response, and this is why people include them in the preview image. And I don't need my emotion to be disturbed.
Having said that, I don't know how practical it would be to identify faces. There's for example face-api.js that can perform face recognition using tensorflow.js core.
The Tiny Face Detector is a very performant, realtime face detector, which is much faster, smaller and less resource consuming compared to the SSD Mobilenet V1 face detector, in return it performs slightly less well on detecting small faces. This model is extremely mobile and web friendly, thus it should be your GO-TO face detector on mobile devices and resource limited clients. The size of the quantized model is only 190 KB (tiny_face_detector_model).
The text was updated successfully, but these errors were encountered:
There are a lot of things some people do not like to see on mastodon. This can be images of food, weapons, naked skin, certain animals or - even faces. This often leads to fights over server content policies and content warnings that are ridiculous, IMHO.
I like the idea of having "client side" filters for these very specific "needs".
Now with powerful image classification systems this is possible.
I want to support this idea in the hope that these kinds of special content-filter "needs" are solely in the responsibility of the viewer and not the poster. 👍
When browsing through the posts I don't like to see preview images of human faces, especially politician faces. Human faces elicit automatic emotional response, and this is why people include them in the preview image. And I don't need my emotion to be disturbed.
Having said that, I don't know how practical it would be to identify faces. There's for example face-api.js that can perform face recognition using tensorflow.js core.
The text was updated successfully, but these errors were encountered: