Working with BabylonJS & AI

AI is huge!

Reference Image from my demo BJS app:

Send to Midjourney with a little prompt

Using Image Weight --iw 2 (to bring the image closer to the reference)

More examples


This is insane I love it !!!

1 Like

This is suuuuper cool! I’ve seen some ideas how a 3D base can make the image generation much more guided, it’s pretty interesting :slight_smile:

1 Like

I guess now the AI just needs to be able to generate all the materials and lightmaps and next I can start packing-up my designer’s gear and plan on growing potatoes in my backyard to make a living. Who’s gonna need a designer after that? Just watch your asses, ya devs, because yours are the next ones on the list :stuck_out_tongue_winking_eye: :crazy_face:

AI is going to kill us all, Elon said :fire:

Except Elon considers himself above us human beings :zipper_mouth_face:

and by “others” I also mean “artificial others” :smiley:

So who the faen is Johann ML Brown? Are there actually people making money, building a reputation or improving their lives in any way by publishing these fine words, …or is it just for fun, self-esteem or scratching their own back/ego? :thinking: :stuck_out_tongue_closed_eyes:

Hey chatGPT, make this shorter: comparing yoursef to others is a fool’s game and will steal your joy. The only person to compare yourself to is the person you were yesterday.

ChatGPT: Comparing yourself to others steals joy. Compare only to your past self.

Me: shorter

ChatGPT: Don’t compare, just focus on personal growth.

Me: shorter

ChatGPT: Focus on self-improvement, not comparisons.

Me: shorter!

ChatGPT: Embrace self-growth, not comparisons.

Me: shorter!!

ChatGPT: Grow, don’t compare.

Me: shorter!

ChatGPT: Grow!

Me to Me: Shorter!’
Answer: ‘…?!’

Sidenote to myself on this last answer:
"Ok, I guess that’s clean, humble and open enough for me :grin: I’ll take that for an answer :crazy_face: :stuck_out_tongue_closed_eyes: :rofl: "

Yeah. MS Bing chat is now available in my country and it looks like it will help me with my usecase:
AI is great and I love things like chatGTP or midjourney but none seemed to be really helpful for ‘executional work’. I mean chatGTP cannot give you Internet links and others are for creating stuff and…
I had this pending very basic request for execution work for the scene I’m building (the space museum).
I need to nurture my interactive sphere main feature with 60 thumbs (of pictures or videos) for each category/tags. I.E. natural wonders (60), monuments (60), pictures representing in each era things like ‘art of war’ (60 per era), ‘faith’ (60 per era), ‘science’ (60 per era)… I guess you got the point. So, I need to fetch a link for all these with respect to the tags above + some tech constraints like (minimum 1k res, landscape only) and some commercial constraits (like free of rights).
So here the good news (I think). From the quick first tests I did using Bing Chat, it seems like it would actually deliver. I’ll let you know later on if this collab turned out to be a success and how much additional efforts I had to put into it…