Turning a chair into a table, or vice versa, might sound like somewhat of a magic trick. In this case, zero magic is involved, just plenty of complex geometry and machine learning.
Called LOGAN, the deep neural network, i.e., a machine of sorts, can learn to transform the shapes of two different objects, for example, a chair and a table, in a natural way, without seeing any paired transforms between the shapes. All the machine had seen was a bunch of tables and a bunch of chairs, and it could automatically translate shapes between the two unpaired domains. LOGAN can also automatically perform both content and style transfers between two different types of shapes without any changes to its network architecture. (1)
Chair… Table… Human… Cosmos…
Look at any shape.
Imagine any shape.
There are ways to go from one to the other. But there is nothing natural about it. All changes are abrupt. Raw. Untamed. Whenever something becomes something else, the first one dies. Completely and utterly. There is no gradual change. No “natural” way of dying. No “natural” way of changing. This is the secret we have chosen to ignore. And we keep on believing in the ability to change. This is the essence of our civilization. The cornerstone of our existence. That we can “change”. That things “change”.
Imagine a cosmos where everything is stable.
A perfect cosmos.
We hate this cosmos. For it nullifies existence.
Free beings we are.
And if we choose, we can choose to be!
And die…
So have we done.
So shall it be…
