Show HN: I built an AI art installation at home generating new pieces on the fly
I just finished my latest project, building an AI art installation at home, generating 100 % unique artworks on the fly. Just push the button below the screen and another one will be displayed. When the button has been pushed, the old artwork is deleted and can't be retrieved again.
Setup:
* An Nvidia Jetson Xavier NX was used for all logic, machine learning inference, art kiosk GUI etc.
* A StyleGAN was used to generate the artworks, trained on ~5k images of abstract art.
* A passive infrared sensor (SR602) was integrated with the Jetson to reduce screen burn-in. When no movement has been detected around the installation within a pre-defined threshold, the screen shuts off until movement is detected.
* A custom control box was built, encapsulating most of the electronics.
This says a lot about building MDF housing for a Jetson (BTW why not plywood?), and devotes literally half a dozen pretty generic lines to the actual art generation code. The algorithm is not discussed, and even art examples are not shown.
This is sad, because wooden box building guides are abundant, but art generation guides are less so,
I wonder if using a color e-ink display would make the art look more appealing on a wall.
Now have it automatically issue one NFT every few minutes. Profit!
Cool! Could also be nice with a A3 color printer and paper shredder. Banksy-style.
This would make it less tv-isih
Would it have been easier to generate these images on some cloud gpu and stream/send the images to a smart tv? To avoid building and fabricating all the hardware components?
Really cool. One piece of feedback: Try to get the white balance to better match the warm wight light of the room. The tv will blend in much much better.
The creator of ElectricSheep installs large installations like this. https://en.wikipedia.org/wiki/Scott_Draves it uses crowdsourced human preference functions to drive the art evolution
Check out 77 Million Paintings by Brain Eno
Max Fischer? By any chance did you petition to keep a Latin class at your private high school? ;)
Very cool. I work from home in my basement and would enjoy such a device. Any plans to sell them? Or know of similar devices? I could probably build one, but most likely would end up in pieces next to my other random projects…
As cool as this is, I have been struggling to find a good non-AI digital photo display for my house.
The common ones you can buy as “digital photo frame” on Amazon sucks in image quality. I would like a minimalistic digital photo display that is colour accurate, hidpi, preferably have a way to store a handful of hi res images(if not, Adding a Pi-zero for this is still okay). Oh I am not so rich or hosting an art gallery or anything, so, it must be less than 700 bucks(the cost of an iPad). I don’t understand why such a thing doesn’t exist already.
As an artist and an avid reader of HN, I find this both fascinating and terrifying (!) at the same time. Before I started painting robots, I painted abstracts and was heavily influenced by Abstract Expressionism. The images that your AI are generating (creating?!) are really cool. Equally as cool is that the images are deleted once a new one is generated, because a lot of the famous artists destroyed or painted over old works.
I wonder if you have to push the button or if it could auto-cycle at some interval. Powering is always interesting do usually need a dedicated power source like a wall plug less you had some other means to get power. Low energy/battery/harvesting maybe (I saw this runs on a Jetson).
Brilliant: the project itself but also the trouble you took to share the how-to deets!
Nice.
What do you think of this? : https://vimeo.com/241051006
Also, here's a generative art discord : https://discord.gg/gVna7Utm
First, I would like to say well done. Not just the project -- it is neat. I particularly appreciate the fact that it includes all the less obvious, to me, pieces like building appropriate box, sanding and so on.
I wish more projects were like this ( as in, showed most of the steps ).
Maybe a stupid question, but why the Samsung connect box if the Jetson already has a HDMI output?
This is awesome. Is the StyleGAN code removed (to keep it private)? I see this commit message: "Remove my ML stuff." I'm interested in understanding the StyleGAN if you are open to sharing it.
Very cool. But have you put it up against the AI Turing test? :) https://gumgum.com/guides/artificial
undefined
I love how this read me is as much about woodwork as it is about software
Very beautiful. If I trained the network on say, just drawings and painting Mondrian did, do you think they'd converge stylistically?
Is a huge (5,000+ items) dataset required to get decent results?
This reminds me a lot of this Japanese Product/Service: https://frm.fm/overview
Non tech nitpick: biscuit joining a box made out of mdf seems like massie overkill when plain old wood glue would have absolutely sufficed.
I was thinking the Frame TV had better color grading for the color temperature of the room. Does your video output not take advantage of this?
The installation is impressive but the output is underwhelming. Then again, I feel the same about the pieces in most contemporary art museums.
Samsung phones come with an adapter that can output to HDMI. Is there a phone app that could do something like this on a TV?
Nice! Art being lost forever is art in itself.
Personally believe that art requires intention and some attempt at communication but it's an interesting project.
Very cool. I now have a reason to get my hands on a Jetson. Also great work on the detailed readme.
Worth considering that you could purchase a very good painting from a local artist or gallery for much less than the money and time that was spent on this project. While I get that this is a fun project, an actual work of art will provide much more enrichment over the many years that it will outlive an AI art installation.
Really fun stuff, perhaps you should use a computer screen instead of producing one
I'm curious about, beyond time and labor, how much $$$ the project ran you.
Looks very cool! I think there may be a demand for that!
a possible next step > add a motion sensor to *swipe fwd/back the artworks. using a smooth pixel morphing blended transition.
*or when toilet flushes ;-)
do you find any of the generated images interesting? there is no thought behind any of them which leads me curious to how captivating they might be.
use CLIP to generate a title for each piece of art, followed by a description with GPT, and it is ready for exhibition
That's extremely cool. Well done. :)
undefined
undefined
This is so freakin cool
This is amazingly cool.
is there a desktop wallpaper version?
I like the idea, but the art is meh.
great write up.
Taking things my kid could do into the next generation. Awesome.
this is just proof that any splash of shit is abstract art
it's funny what tech people consider Art. LOL
I hope you aren't syncing these random images to your iCloud account!
Art shouldn't be unlimited. Its ok to call things screensavers, because screensavers are also cool, its just that it isnt art in the sense of "art" the word, if an AI is doing it, its something else, and making it unlimited is not interesting, it is irresponsible. Its the constraints that make art what it is, not simply the definition of an infinite possibility space
