Show HN: Lil digi – play a platformer game as yourself
lildigi.meI've been monitoring some of the image generation progress for game sprites for a while and was really impressed by Gemini Flash. This demo lets you upload a picture and turns it into an animated pixel character (a lil digi gif that you can download). I chain together the pixel art frames and insert the player into a tricky platforming level.
video demo: https://www.youtube.com/watch?v=EK7z-HltssE
api usage: Unfortunately Gemini Flash 2.0 image Gen has really tight API limits that don't scale up for paid tiers -- if anyone on the team sees this and wants to help me out, my email is in my profile or you can reach me at hello@lildigi.me
level stats: For the speedrunners out there, the fastest glitchless times I've seen around 17-18s.
stack: cloudflare pages + aws lambda + gemini
happy to answer any questions!
This is one of the most delightful things I've seen on the internet in a long while. I immediately started sharing it with my friends and we shared the cute little characters we made. Everyone loved it.
Thank you, I appreciate you sharing that!
That's a fantastic idea! I love the little me on that - taken from just a webcam pic with a messy background and it's pulled out things like my tshirt perfectly.
Thanks! A few friends have starting using the generated digis as profile pictures because of how recognizable they are. I was definitely surprised at how good the models had gotten.
Was the general art of the game made by an AI like PixelLab?
I used gemini for the general art as well!
Very neat. Would you share some of the prompts you used to generate the pixel art sprite? Super accurate to the photo I provided.
Thanks! Like another user commented below, I think the appeal of the pixel art sprites is that at a lower level of detail it can feel really accurate. 10 different real life red jackets can all end up as the same pixel art representation but each person would recognize it as their jacket! It feels like a form of compression.
On prompting, you can get most of the way there in AI studio on Gemini 2.0 Flash (Image Generation) Experimental by uploading a picture and asking for "a high quality detailed pixel art sprite of this character." Most of the backend annoyance here was iterating to improve prompt adherence (characters not facing the same way, outfits changing between frames, etc).
I am extremely impressed. I literally put my hand fully in front of my face, and it got me spot on. My glasses were partially shown, as was my beard and hair.
That being said, the resolution is such that saying messy hair, full beard, and black glasses, would pretty much get it.
He's probably feeding the webcam pic as img2img no? That'd get you the most details from the original picked up by the AI
I uploaded a photo, though. I didn’t give permission for the camera.
I think it’s simply a matter of it assumed a person, found glasses, and a bit of hair above and below and filled in logically.
thanks for the traffic everyone -- definitely hitting rate limits on gemini already so if it doesn't work please give it a try later in the day or try the example.
Apologies for the inconvenience! If anyone on the gemini team can help out, please lmk (email in bio).
THIS IS SO COOL, got top 6.9% haha https://imgur.com/a/8MptH2l
nice! I was personally surprised seeing scores roll in that were lower than mine, since I had played the level so many times while debugging haha.
Best in the system now is 17.1s and it must have been a near perfect run.
Brilliant idea and great execution!
thanks for the kind words!
Great work!
damn such a cool idea, didn't let me play as donald trump lmao but this concept is so cool.
Thanks! There are some content filtering rules via Gemini, but if you got an error on Trump I'm guessing it's an issue with the image generation rate limit getting overloaded.
Man you motherfuckers are smart.
This is pretty cool