Hi everyone! I'm Yuri, and this is my second project for the Easy Lens Challenge by Contra!

When I was working on my project for this contest, I fell in love with the idea of creating a mini-game controlled entirely by head movements. Since Easy Lens is fantastic at face recognition and tracking position, that’s exactly how Space Defender - Save the Galaxy! was born.

I started by formulating the idea, mapping out the functionality, game logic, and everything that should appear on screen. To bring it to life, I used Claude to speed up breaking down the overall structure and crafting the final prompts. Then, I moved on to generation with Easy Lens. Just like with my previous project, my strategy using Easy Lens was to get the core game logic working first and refine the visual elements afterward.

Check out how it turned out! Step into the shoes of a real Space Defender wearing a cool helmet. Blast monsters with a laser—to fire it, simply smile wide or open your mouth! Aim using your head movements, but be sure to avoid the bombs! Complete all 3 levels, stretch those neck muscles, and save the galaxy!

I hope this mini-game brings a smile to your face as well.

🔗 The result of my work in Easy Lens: https://www.snapchat.com/unlock/?type=SNAPCODE&uuid=e3506509a51e47838d08ceb2277d6d21&metadata=01

🔗 Post on  X: https://x.com/YuriiFunkendorf/status/2083159438555238854?s=20

Thank you for your attention! See you!

Amazing job, Yuri! Massive good luck in the Challenge! 🛸

Can the same prompt start feeling like you? 🥸

I'm not a fan of long, super-specific prompts, and much more interested in building systems where my vision lives naturally in every single image. So I'm pretty picky about creating personal styles for each of my projects

Sometimes one sref is enough, sometimes it takes a combination of a few, and sometimes I need a whole moodboard

Personalization used to feel like the most inconvenient option because I had to go through hundreds of image pairs, which took me daaaays. But apparently the whole thing has been redesigned in 8.2 and now all the images are shown in one even grid, and you need to choose waaaay fewer to get a really good result!

So in case you've been avoiding Midjourney personalization too, you can probably stop now 🎉

✦ In the video: my style for aftrglo compared to the default generations Love it sooo much, my current favorite❤️‍🔥

Drop a prompt in the comments (even one word hehe) and I'll send you the image it creates🙌

The 8.2 grid change sounds like a genuine quality of life fix, hundreds of pairs down to one grid. Curious what tips you off that a single sref isn't going to cut it and you need the whole moodboard instead, is that decided project by project or do certain aesthetics just always need more scaffolding?