Google Inceptisism - Machines That Can Produce Art and Even Dreams

Who's Online?

We have 737 guests and no members online

Tuesday, May 21, 2019
Home  >  Beams of light  >  Spectrum of lights  >  Technology  >  Google Inceptisism - Machines That Can Produce Art and Even Dreams

User Rating: 5 / 5

Star ActiveStar ActiveStar ActiveStar ActiveStar Active
 

Internet Marketing Strategy

 

The brain of the internet, has recently demonstrated that its artificial neural networks have the power to dream by trained these networks to recognize objects, animals, faces and nature, by feeding them millions of images. These networks have become so advanced that they are able to "enhance" what they are seeing. The images produced from these enhancements have been dubbed, "Inceptionism".

 

thescream-mod3

Edvard Munch "The Scream" After Going Through Inceptisism

 

These artificial neural networks are structured in stacks of up to 30 layers. The bottom layers detect low level information like edges and corners. Going up, every layer fills in a little bit more information until the top layers of the network determine what is being seen.

 

Computer_Art_1

 

'If we choose higher-level layers, which identify more sophisticated features in images, complex features or even whole objects tend to emerge. 'Again, we just start with an existing image and give it to our neural net. The team found this creates a feedback loop. If a cloud looks a little bit like a bird, the network will make it look more like a bird. This in turn will make the network recognize the bird even more strongly on the next pass and so forth, until a highly detailed bird appears, seemingly out of nowhere.

 

Computer_Art_2

 

'The results are intriguing—even a relatively simple neural network can be used to over-interpret an image, just like as children we enjoyed watching clouds and interpreting the random shapes. 'This network was trained mostly on images of animals, so naturally it tends to interpret shapes as animals. 'But because the data is stored at such a high abstraction, the results are an interesting remix of these learned features.'

 

Compouter_art_4

 

In this picture, the bottom layer of the network responsible for detecting edges and strokes was asked to build upon what it was seeing.

 

sky_bright

 

In this example, the neural networks were trained to detect animals. They started with an ordinary picture of a cloud against a blue sky and the network enhanced what it perceived to be animals.  

 

Some Close-Ups From the Last Picture

CUP_Computer_Art_animalsfunny-animals

 

 

red-tree-small-long-unsmoothed

 "Whatever you see here, I want More if it!"

 

Artificial neural networks are a form of artificial intelligence based on biological neural networks, such as the central nervous system. Unlike more conventional software that works according to rigidly defined rules, artificial neural networks are trained by being shown millions of training examples and having their network parameters slowing adjusted until they return the desired results.

 

Comoputer_Art_5

 

The network is made up of nodes that act like interconnected neurons. These allow computers to handle large numbers of data inputs, the nature of which are either unknown or approximate. The idea is that these nodes are set in stacked layers with each layer handling data of increasing complexity, which they weigh according to their instructions. These layers have the ability to learn and become better at detecting desired patterns as they are fed more data over time.

 

Computer_Art_7

 

This ability has made artificial neural networks very valuable in areas like image classification and speech recognition. For example, image recognition systems using artificial neural networks can handle images in the real-world chaos of unpredictable lighting, angles, colors, and backgrounds. Instead of following strict rules for sorting out what in a picture is a banana, a neural network can operate by using millions of examples to learn how to find simple things like edges, combine these into more complex shapes, and then combine these to identify various objects – one of which could be a banana.

 

seurat-layout

The Google research team sees Inceptionism as having a number of applications beyond the psychedelic. In addition to a better understanding of artificial neural networks, the technique can help to improve network architecture and as a check on network learning. It may even one day become a new tool for artists and even provide new insights into the creative process.

 

  

 

Ostagram is based off DeepDream, a software pioneered by Google under the codename Inception and the obvious inspiration for the art style's name. What this means is that Ostagram finds patterns within images and attempts to bring these patterns together into a resulting image–a process referred to as algorithmic pareidolia. Based on an art style known as Inceptionism, these photos are the result of two images being combined together using neural networks, a computer system designed to mimic the activity of a human's brain, and the underlying foundation of many deep learning algorithms.

 

Russian Website "Ostagram" Uses Neural Networks

to Combine Images with Awesome Results

 

ostagram-russia-neural-networks-art-inception

 

A Sci-Fi UFO landscape, Styled after Van Gogh

 

combinacion-imagenes-red-neuronal-13

 

Despite how it may look, the above images are not a strange closeup of a Van Gogh painting. No, they are creative and bizarre creations developed by a user of Russian web site Ostagram using the principles of Google Inceptisim.

 

combinacion-imagenes-red-neuronal-10

Leonardo diCaprio/Oscar -  Best Actor 2015 for "The Revenant"

 

Based on an art style appropriately referred to as Inceptionism, this photo is the result of two images being combined together using neural networks, a computer system designed to mimic the activity of a human's brain, and the underlying foundation of many deep learning algorithms.

 

Ostagram_Arnold_Swhartneger_1

Actor and Former Governor of California Arnold Schwarzenegger

Ostagram is based off DeepDream, a software pioneered by Google under the codename Inception (a nod to Christopher Nolan's mind-bending movie) and the obvious inspiration for the art style's name. What this means is that Ostagram finds patterns within images and attempts to bring these patterns together into a resulting image–a process referred to as algorithmic pareidolia.

 

combinacion-imagenes-red-neuronal-4

 

What's 'algorithmic pareidolia'? According to DeepDream's Wikipedia page, it's 'a dreamlike hallucinogenic appearance in the deliberately over-processed images.' In other words, it's an acid trip brought to life.

 

combinacion-imagenes-red-neuronal-16

 

Below are additional trippy images created using "Ostagram Neural Networks" photo combinations:

ostagram_black_cat

 

combinacion-imagenes-red-neuronal-2

 

combinacion-imagenes-red-neuronal-3

 

ostagram_1see_and_tower

 

combinacion-imagenes-red-neuronal-5

 

combinacion-imagenes-red-neuronal-6

 

combinacion-imagenes-red-neuronal-7

 

combinacion-imagenes-red-neuronal-8

 

combinacion-imagenes-red-neuronal-9

 

combinacion-imagenes-red-neuronal-11

 

combinacion-imagenes-red-neuronal-12

 

combinacion-imagenes-red-neuronal-14

 

combinacion-imagenes-red-neuronal-15

 

combinacion-imagenes-red-neuronal-17