Kiki Wu
World-building, and creating surreal scenes for virtual events and exhibitions have been crucial parts of my artistic practice, as it enables me to bridge real-life experiences with the digital realm. This process of creating digital art in cyberspace is integral to my work, as it allows me to translate tangible moments and emotions into immersive experiences.
In this article, I will outline techniques for crafting immersive, surreal scenes using the latest AI image generators and VR platforms. The techniques presented are beginner-friendly and do not require any coding skills!
Although immersive artwork is commonly associated with virtual reality nowadays, the concept of immersive experiences can encompass any type of experience that fully engages the viewer's senses. This can include XR artworks or large-scale interactive installations like what Teamlab is doing, which transports the viewer into an entirely different world or environment, where they can interact with the artwork/content using all of their senses, not limited to visual senses. Here I will concentrate solely on art experiences that are based on virtual reality and web browsers.
Research has shown that humans tend to learn better and retain more information in immersive environments. This is because immersive experiences are more engaging and memorable, allowing the viewer to fully interact with the content and absorb the information more effectively. This makes immersive art an increasingly important tool for education, especially in fields such as science, history, and the arts.
The global pandemic happened in 2020 may also be a contributing factor to the growing popularity of immersive artwork. With physical exhibition venues being shut down, artists have had to find new ways to present their artwork to the public. XR experiences that can be hosted online and viewed repeatedly have increased the accessibility of art while also promoting the healing effects of immersive virtual worlds that can help reduce people's anxiety during long periods of lockdown. This has made immersive artwork a particularly attractive medium for artists seeking to engage with audiences in new and innovative ways, while also providing a means of artistic expression during times of social distancing and isolation.
Creating immersive artwork often requires possessing 3D modeling skills, as well as knowledge of game engines such as Unity and Unreal, or WebVR library like A-Frame, which are used to set up the virtual reality world and create interaction within the artwork.
Fortunately, in 2023, artists who lack coding skills can still create immersive scenes using metaverse platforms such as STYLY, Spatial, they are mostly web browser based editors which enable creators to creating immersive artworks by simply uploading 3D models and other media.
Moreover, with the increasing popularity of AI image generators and ongoing improvements in this technology, artists can easily draft concepts for their artwork and build beautiful prototypes using these tools.
Go to Blockade Lab and type in prompts to create AI 360° panorama image. We will use this images as the main scene in our immersive artwork.
It will take around 15~20 seconds to generate the image.
💡Here is my prompt: Inside mystical mushrooms, curly vines and blue lotuses, psychedelic flowers, trippy neon color, floating stars.
You have the option to choose from 16 different art styles. I prefer the Anime Art Style, Dreamlike and Surreal, as they align with my affinity for psychedelic art, trippy color schemes, and gradient visuals. However, feel free to explore and experiment with other styles and type in extra style words until you find the one that resonates with your unique creative vision.
For your reference, take a look at these mushrooms scenes of different styles:
💡Struggle with prompts? Don’t worry, you can find inspirations on PromptHero
Don’t like art style the built in Blackade Lab? Let’s jump to MidJourney.
Pay attention to the prompt here:
a seamless equirectangular 360° panorama photo of Your Imagination, Art Styles - - ar 2:1
The first part “A seamless equirectangular…….” is crucial because it defines the image type and format, and the last “- -ar 2:1” is the aspect ratio of your image. Remember to include these two parts in your prompt!
💡New to MidJourney? Check the user guide here!
After you receive the output from MidJourney, download the image and upload it to Blockade Labs to check for any potential bugs.
It's important to note that while you can input prompt like "seamless," MidJourney is not specifically designed for generating 360-degree panorama images. As a result, there may still be visible seams in your image even with that prompt.
So let’s fix the seam! Go to photoshop and import your AI image, cut the image in half and switch both sides. Move the right side to the left and the left side to the right.
Your resulting image should look like this:
You see seam! Let’s get rid of it using Spot Healing Brush tool (Shortcut is J)
Now your AI image is ready! Hit the File > Export.
In order to make the 360° panorama image the primary scene, I will make it as the texture for a skybox object.
A skybox object is a technique in computer graphics that creates a realistic outdoor environment, like a sky or distant scenery. To create this effect, I wrap the AI image as image texture around a sphere in Blender and export a .glb file, upload it as a skybox object to STYLY.
Luckily, the social VR platform Spatial.io offers a skybox template that streamlines the process and eliminates the need to use Blender! By following the provided steps, creators can easily generate their own skybox without any hassle.