Hi! My name is Eva, and I'm writing about using data in art. The article is about how data plays an essential role in making art. It touches on deep learning and synthetic images like Generative Adversarial Networks. To summarize this incredibly complex subject into a unique experience. It's best to showcase what can be done with data from your own perspective.
The use of data in art is a hot topic in the art world. We have so much information, details, and figures about the market and the people. Artists, in a way, have to know what kind of tastes their audience has so they can adjust their work accordingly to this data, or they will not succeed. At least, that's how it works for conceptual artists who seem to be aware that the work has been consumed.
This is a process of teaching computers to learn from data without being explicitly programmed. Instead, machine learning focuses on developing computer programs that can access data and use it to learn for themselves.
Deep learning is a subfield of machine learning that is a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers composed of various linear and non-linear transformations.
Neural networks are computer systems that are designed to recognize patterns. They can be trained to perform image classification, speech recognition, and machine translation. They are also being used to generate images, which is what we call neural network art. You can train a neural network to generate a specific type of image. For example, you can train a neural network to generate images of cats. Then, if you show it a random image, it will try to generate a cat image.
Generative Adversarial Networks
A GAN is a deep learning algorithm where two neural networks, called generator and discriminator, compete against each other in a zero-sum game framework. It is used to generate synthetic data.
Synthetic data is engineered to be more diverse than real-world data. It is highly randomized across a broad range of parameters like lighting, camera angles, and object placement to prepare your model to handle any condition it might see in the real world. Plus, images come pre-labeled and annotated, reducing the potential for human error.
Data art is closely related to generative adversarial networks (GANs), which are essentially a form of machine learning that trains a model to generate realistic images. An intriguing example is data sculpting.
"Data sculpting is a creative process where data is the material out of which the created works take shape" by Dariusz Gross.
I'm fascinated by the possibilities of using data in 3D by Dariusz Gross. This is a fantastic case of data sculpting. The example below is an image GAN from one of his recent works called "mama tata," created using deep learning code.
I believe data is the future of art. I also think that data can be used for human intelligence enhancement. However, it should not be used just as a material for art to achieve goals but also in data science. That's why I'm writing this article.
Be sure to SUBSCRIBE here to never miss another article on digital art 🟣
Artists and Engineers are working together, and I think there will be more collaboration in the future because we have much to gain from each other.
Dear Eva,
I have been following your interesting medium-blog for some time now and I am a fan of it. This article is very interesting to me as I am also interested in how we can use data in art. It is amazing that artists nowadays realize the importance of data and start working on creative efforts based on it. It is important that artists face the reality and then work to make changes to the world around them by using data to transform the way we live.
I like the article. I look forward to seeing how the article goes.