In a recent post... I had a bit of a ramble about how, for a very long time, I've always thought that the Florence and the Machine song, "Caught" would make an excellent series of works, but it has a very, very abstract and metaphorical lyrical tone throughout it, that does not lend itself well with direct representation.
My history in this realm is well versed in traditional artistic practice:
Converting song into visual is something that has constantly fascinated be as a creator. My first major piece of work at high school (Go read about how I ended up with a Master's Degree in Visual Art) was inspired by song, and I would go onto to use music as a major influence in my work, all the way through Photographic outputs, where I took a "playlist" of favourite songs and completed a portrait image of each, accompanied by the lyrics.
I hope to produce more in this style, and there's a certain piece of music that has haunted me in terms of its powerful representation and narrative that I simply just must produce something to get the imagery out of my mind. More to come on that one. Possibly in the next 48 hours!
Caught (Frame 1?) (very much a work in progress)
It just gains its strength and digs its hooks, To drag me through the day
The prompt itself was
defeated female warrior, dragged across battlefield, ghosts, sickle hooks, highly detailed, dramatic, dark, 8k, realism,
The entire song is something I want to represent in a series at a - it is a strong source of visual metaphor with very powerful imagery that is just dying (in my mind) to emerge from the void into something more coherent than the image above.
I have way too many elaborate notions, influenced by this song. It's a beautiful read as lyrics alone, and an even more beautiful listen.
At the same time, there is a certain danger in allowing a concept to become unleashed and let itself out open the world. Within my mind, I feel pangs of regret at unbottling that potential. It is a kind of writers block, an empty page is untarnished, but the moment you write a single word upon it, you've ruined it with your mortal form.
None of this, of course, really matters all that much, so I may as well develop this creation, even if the the song "Caught" by Florence and the Machine starts with a phrase that loops around the very rambling concept I've just covered.
It's the hardest thing I've ever had to do
So here we go.
Definitely needs a path to be depicted. I thought that frame 3's buildings looked somewhat like a forest. It's not the case, they're in fact, buildings. So I tweaked the prompt:
the hardest thing I've ever had to do, dark, lush green forest, with ruined arches --ar 16:9
The lack of a character in this composition takes away the sentiment of the original line, so It will be unused, and I will change the prompt again, adding the term "a woman" to the above one.
The outcome is slightly more promising:
However, I do not see a "woman" present in any of these compositions with the except of 3 and 4. This will simply not do. Maybe If I put in the entire lyrics of the song?
Certainly not. These are just abstract landscapes.
I'll try then to combine the first two lines of the song:
It's the hardest thing I've ever had to do
To try and keep from calling you
I feel as though this a cosmopolitan scene, and to represent the telephone, what better way than to use a phone booth? Here goes nothing:
close up of a sad, beautiful woman contemplating a telephone box on a desolate city street with windswept mountains in the background behind the buildings of the city at golden hour --ar 16:9
Is.. that the Samsung logo on number four? :D Possibly. I like number four, but it doesn't feel "British" enough given the iconic red of the phone box.
/imagine prompt: street portrait extreme close up of a sad, beautiful woman contemplating a telephone box on a london city street behind the buildings of the city at golden hour --ar 16:9
This is pretty abstract. Particularly, number three, where the woman is trapped within the phone booth. I want things to be a bit more literal than that in terms of representation. I'm starting to realise that now.
To this point, I've just used MidJourney on this, so I want to see what Stable Diffusion, running locally, can do with the same prompt...
Super bizzare, but I'll let it have some other attempts as well...
Not sure about Sable Diffusion here - truly bizarre faces and really interesting telephone box names. The third frame is pretty good, I think I could potentially work with that random seed, but it will be difficult to get stylistic consistency throughout the series if I am swinging wildly through the very.. varied lyrics of the song.
Simultaneously, I wonder whether I should be even attempting to push this song into a visual. It again feels as though by working on such a thing it may just destroy the original beauty of the song and what attracted me to it in the very first place, the enigmatic nature of the lyrics, and the powerful visions it evokes within me.
Having said that, nothing good comes out of not doing anything. I have to keep trying, and have to keep working away at this notion. I will represent this in a literal, or metaphorical sense, one way or another, but it is clear I will not get there today.
That is okay. Art is not always about succeeding on your first attempt at something.
Want more content from me?
Witness my futile efforts to play my Steam Game collection in alphabetical order.
Are you aware that I love photography? Check out my work in a collection.
If you want to see my Splinterlands antics and rants live, Find me on Twitch
If you prefer sleeping in your designated time zone, go watch replays on YouTube.
In some ways AI art pushes us in new directions; but we need to remember AI is only showing us the ways those prompts/concepts normally relate to each other.
It has to 'draw' from somewhere, obviously; but drawing from the human experience is only going to reflect the human experience.
It just shows us things that weren't an obvious connection to our feeble man-brains. The new models that are being trained are specifically avoiding any new AI created materials, to ensure that humanity continues to be the base level inspiration for this stuff.
Really cool concepts! Yesterday, I've prompted the lyrics of my second album, and got results with very similar colors as this one:
I am working on a hyperealist portraits post, but here is one my pic from my album:
Similar colours right?
Fascinating. It seems to have a vernacular of its own that seems to gel with this style of prompt. There's simply too much to explore in this field, and I have dozens of experiments that I still want to run!
I agree with ya a million! Not a single without me having an idea to prompt, it's getting crazy!!
Anyway keep up the great work!!!
That first image sending shivers thru my spine.
It seems to have some hard-to-tell stories.
Is this part of some series?
Definitely much more to come. I don't think I'll ever be happy with the depictions I've developed / am working on related to this song and theme.