Phase 1: Translating hypnagogic imagery using txt2img AI models
In this Phase I am investigating how AI models like Dall-e could be used to translate the hypnagogic imagery that I have into a visible visual image. For this I go into the hypnagogic state, find a vision and create a text prompt based on the same. This is then entered to a txt2img model to convert into images. I then curate and edit it to reach the result I feel best. More often the results are not a direct 1:1 from the hypnagogic images. This is because I feel that once I receive the hypnagogic image, my logical brain should interpret it in the way it should, which will give the best results. Thus, ensuring a left brain - right brain collaboration of sorts.
1. "Beyond Immersion"
One of the first image that I received the hypno state was that of a woman (maybe Indian) with a hooded scarf wearing a VR headset. This suddenly evolved into multiple characters with VR headsets. Finally taking it to my logical brain, I found this particular image to fit what I wanted to share. Here the Buddha wears a VR headset and it became a symbol for the intersection of 1) the the external world "the Universe", 2) the interconnected VR world "the Metaverse" and 3) the internal hypnagogic world "the Numaverse". The title is "Beyond Immersion" as I (and we all) would and should move beyond the immersion of all these 3 worlds. And beyond the immersion I feel all 3 worlds would converge into one unity. But this for later.
Connection: An interesting occurrence which happened when I released this on social media was that my cousin brother said he had seen this exact same image in a dream many years back. He was genuinely shocked and happy at the same time to see the image.
Second Connection: My initial iterations for creating the above image included these variations:
More than a month later, I saw this image in my twitter feed created by the user "The Imaginary Teapot":
This was a non-obvious image and seemed like a very interesting connection.
2. "Listening in Silence"
Inspired by my previous collab with George Boya Formless, I tried to expand the concept of unity-duality. This time around I tried to generate the image first (above) and then fed it into the hypnagogic state to retrieve another image. This was then worded in the AI model to generate an image. The image I retrieved is:
3. "The electric hand"
This image came to me during a meditation and someone close to me was getting a tattoo the same day. I knew about the fact that the person was getting a tattoo in their arm, so this is not necessarily predictive but definitely a connection.
4. "Deep down under the ground"
I went a little more deeper this time and retrieved an image where a slab was kept vertically on a flower pot. Although this was the seed image, I used the text prompt reading this image myself and generated this image using an AI model:
I am not sure if this is a connection, but many days after I created the above image, I saw the below image on a wall as part of an advertisement of a restaurant. Somehow it reminded me suddenly of the above image and I took a photo of it.
5. "The spiral"
Up next I saw an image of a whale going in a spiral in an ocean. I did a quick iteration of this and generated a bunch of image of which this one stood out.
I think this project has a lot of "not controlling" but "letting free" in it. The project is an act of realising images and generating them with an element of curation which leads to an open-unique-new expression, rather than a controlled 1-1 experiment.
This one was different. This one came during my meditation in a near deep-sleep stage. I saw the image of a person with multicoloured energy lines zig-zagging through their body. I fed this into the AI model and gen-curated the following image.
I will from now on, once in a while, be using an "intent" while going into mediation. Will see if I could generate an image on a particular need/theme.
7. "Where we fly together"
This time I decided to use an "intent" before and during the hypnagogic state. The "intent" was: "an image from the future". I then fed the image I saw as a prompt into the AI model. I had to do a bunch of edits in Photoshop to get the image right.
There were multiple possible connections for this image shown below: 1. Image generated by Braulio Balthazar , I was travelling to Singapore the next day after creating this image. So there were many structures like this, like 2) the Apple store (image by Choo Yut Shing, 3) the Marina hotel and artscience museum (from CondeNast traveller), or 4) the gardens by the bay. I have seen these places before so could be a latent memory as well.
In my flight to Kuala Lumpur yesterday (6/10/2022), I had a sudden vision of "a building in the shape of a japanese hand fan". I used it to query an AI model and generate this image.
This was my query screenshot:
Today (07/10/2022) I was in Kuala Lumpur when I saw the Mandarin Oriental Hotel for the first time ever. And I could see an instant connection:
This could either be predictive or I might have had seen the logo before on a building and had it in my subconscious. Either way, the possibility of a Japanese hand fan on a building was not so obvious. Maybe I should look for non-obvious images.
9. "The Mask"
This time I tried if someone I know could do the same. So a person I know received an image in a semi-sleep state. The tried to describe the image in one line as "tribal mask man with a golden body on a black metal bar". I prompted the AI model with the same and generated the following image:
Maybe a connection but the person saw a keychain with the mask from the movie "The Mask" the next day. Also interesting to note is the person finds this image repulsive/scary.
10. "In conversation with That"
This image was created with the intent of "a future impactful image". I am trying to test something called resonant tuning. Details on that will be revealed soon below.
11. "Silence of The Lamps"
So as I mentioned previously I'm trying something new: “Resonant tuning” is where i tune towards an intent and match frequencies/resonate. At the point I match frequencies, an image reveals itself in the hypnagogic state and I create it. The below image is one which i created similarly with the intent "a relevant image from tomorrow". It is a bit hard to let go and focus at the same time in the process but let's see where this leads
12. The hug of the Dervish
I tried a few images in waking state visualisation, where I don't necessarily go into a deep hypnagogic state but visualise in the active waking state. The idea here is "not to force", but let the image come.
13. Setting sail
Boats have always fascinated me. There was a certain period in my life where all my dreams were about boats. There was a flood after this near my home and the place were full of rescue boats. It was probably since then that I was fascinated by dream imagery. Continuing to explore with waking state visualisation here.
14. Connection of the main image (2023)
After about a year from the creation of Numaverse-1, I started Numaverse-2. Coincidentally, i found a very good connection between the main image (left) of Numaverse-1 : The painting (right) titled "PRIMATE" of the amazing artist Ralph Khoury (from 2020).
The connection is either a latent memory, whereby I might have seen Ralph's image sometime back before I created the image, or it could be an inexplicable connection. Will explore more of these in Numaverse-2.