Stable Diffusion mindfulness meditation wellness

The featured image for this post was prompted using Stable Diffusion 1.5, or DALL-E, or Midjourney. All of these are AI models that allow you to turn your words into visual compositions. How great is that?

Using Text To Image Generation As A Mindfulness Exercise

by

Music has healing power and is a fundamental part of my life, so I like to pick a “soundtrack” for every post I write. You are welcome to tune in with it and me 🙂

Have you ever set a goal powerful enough to lose sleep but didn’t know where or how to start?

Even with a strong will and clear objectives, it is sometimes difficult to establish a structured step-by-step plan towards a desired outcome.

Mindfulness, a meditative technique that encourages us to concentrate on the present and observe our thoughts, can help us gain insights and draw a roadmap. However, not everyone is inclined to traditional meditation.

Some people are more art-oriented and might find it easier to meditate while creating something. Nowadays, we can use words to create art thanks to generative AI –how beautiful and profound, considering humankind has come so far because of collective knowledge transmitted by language–.

So, let me show you this mindfulness exercise I’ve been doing with generative A (specifically, using Stable Diffusion to produce artwork). But first, if these terms are new to you, let’s do a quick review for you to have a starting point.

What is generative AI?

Generative Artificial Intelligence (AI) is a category of AI systems that can autonomously generate new content or data based on patterns and knowledge learned from large datasets.

Here’s how it works:

  • First, generative AI uses machine learning algorithms and neural networks to analyze and understand vast amounts of existing data, such as text, images, or music.
  • Then, it leverages this understanding to create new content that aligns with its learned patterns and styles.

For instance, Stable Diffusion is an advanced deep-learning model that can generate images from text descriptions. This is extremely useful in visual arts because it helps artists achieve better results based solely on a mental representation.

Also, generative AI, such as Stable Diffusion, is a powerful tool that enhances the creative process and can aid self-knowledge. You will see why below.

How to use Stable Diffusion?

To get started with Stable Diffusion, follow this simple step-by-step:

  1. Register for free at Playground.ai. I recommend this website to work with Stable Diffusion because you can generate up to 1,000 images daily at no cost.
  2. Click on “Create” at the top bar.
  3. You will see two options: “On a Board” and “New Canvas File”. I prefer “On a Board” because you can upscale images up to 4x, which means you will get a higher resolution and better quality.

Once in your new Board or Canvas, you will see a Prompt box on the left sidebar. Here, you will be asked for your intention. In other words, what do you want to see?

screenshot of Playground.ai interface

A prompt can be a single sentence or a chain of keywords separated by commas. This example will give you a better idea:

A lovely couple having a picnic in the Moon, black and white still, digital Art, perfect composition, beautiful detailed intricate insanely detailed octane render trending on artstation, 8 k artistic photography, photorealistic concept art, soft natural volumetric cinematic perfect light, chiaroscuro, award – winning photograph, masterpiece, oil on canvas, raphael, caravaggio, greg rutkowski, beeple, beksinski, giger

Here is the result:

screenshot of Playground.ai interface

Sometimes, you might want to repeat keywords to put “extra weight” on the instruction. This makes the model prioritize a specific trait. For example, including subjective terms like “beautiful” several times can improve the outcome.

Below the Prompt box, you will see another box where you can exclude details from your image.

screenshot of Playground.ai interface

For instance, you might want to avoid the following:

  • Blurry.
  • Low res.
  • Text.
  • Tow.
  • Disfigured.
  • Bad anatomy.
  • Logo.
  • Wordmark.
  • Writing.
  • Heading.
  • Signature.

On the right sidebar, you can modify the image dimensions, the prompt guidance, the quality and details, the seed, and the sampler.

screenshot of Playground.ai interface

Some of these parameters are technical, but there’s no need to worry about them. Just try different settings and find what works best for you.

How to write a mindful prompt

Prompting is nothing more than giving instructions to an AI model. Now, how do we ensure these instructions will be correctly understood?

Here comes the mindfulness exercise. To write an effective prompt, you first need clarity on what you want. Ask yourself some questions:

  • What is on my mind that I would like to represent visually?
  • How can I express this mental representation using words?
  • What are the specific traits that make this idea unique? (colors, shapes, textures, lighting, and so on).

Write down your first prompt the way it seems more natural to you. Try to be as intentional as possible, choosing every word with purpose.

Also, remember that sometimes less is more. AI will perform your instructions literally, so this is about being clear more than a scholar. For example, if you want an illustration of a black cat sleeping, an effective prompt would be: “A cute black cat sleeping.”

Simplicity is powerful.

screenshot of Playground.ai interface

What if you don’t get the expected result? That’s okay. You’re just playing, learning, and improving.

Actually, if you don’t get what you want, that’s great. Because knowing what you don’t want is a fundamental step in finding out what you want.

So, re-engineer your prompt based on what you get, and have fun with AI.

Benefits of generative AI as a mindfulness tool

The process of turning an idea into visual art can be incredibly enlightening.

First, it is necessary to set a goal. Next, we must translate it into a sequence of structured thoughts (prompt). This practice has helped me be more precise and practical in other areas (e.g., decluttering my wardrobe for a more minimalist lifestyle).

Likewise, it’s fascinating how changing the word order in a prompt or using synonyms can significantly alter the outcome. This happens in life, too: how we say something is as relevant as what we say.

Thus, this mindfulness training can help us improve our communication skills and know ourselves better through self-exploration.

Hopefully, you will find this exercise helpful. I encourage you to give it a try for a couple of days and see what you can learn about yourself. You might be positively surprised at the end.