KIRI Engine - 3D Scanner App

A new #Blender version has been released and there's a lot of good talk about optimizations and speed ups 🏃‍♀️ But why has no-one mentioned we can turn our 3D scenes and models into cartoons so easily now 🤩 Grab the free demo file to try it out!

3D Gaussian Splatting inside Godot🎮 Instead of forcing splats into a mesh pipeline, this project adds a separate rendering pass🤩 and blends it after mesh rendering. And yeah, it’s fully open source🥳

@andrew_v10209 Ooooh if you make something cool, share share share 😁 Always up for seeing NPR renders

@KIRI_Engine_App This just gave me a great idea

A new #Blender version has been released and there's a lot of good talk about optimizations and speed ups 🏃‍♀️ But why has no-one mentioned we can turn our 3D scenes and models into cartoons so easily now 🤩 Grab the free demo file to try it out!

@KingOwenFYI It's super fun! And surprisingly fast

@KIRI_Engine_App This looks like such a cool feature.

@pwebb2006 🤩 It's awesome! Loads of other cool demos there too

@KIRI_Engine_App Very cool 👌👍

@Almorgand Always welcome man 😁

@KIRI_Engine_App Came to KIRI for the scanning Stayed for the blender tricks! Thanks for sharing.

We removed the export limit🤩 PhotoScan now has unlimited scans and unlimited exports!!! The first question people asked was: “How do you guys make money then?🤔” Simple answer: we don’t believe basic features should be paywalled. Unlimited should just be normal. What people actually pay for is innovation like 3DGS → Mesh and AI-enhanced LiDAR📹 Make the basics free. Build new things on top.

Scanning glass or porcelain usually means two things: Ruining the object with messy sprays and dusting. Buying an expensive polarized light setup. 😭 We built Featureless Scan into KIRI Engine to skip all that. No spray, no expensive gear. Just clean scans of shiny surfaces. ✨

Mind blown by this new CVPR 2026 paper: tttLRM🤯 It introduces a Test-Time Training layer for long-context autoregressive 3D reconstruction with linear complexity. Multiple image observations are compressed into fast weights that keep updating during inference, forming an implicit scene representation that can be decoded into explicit formats such as 3D Gaussian Splats. Progressive 3D reconstruction from streaming inputs is getting wild🥳It’s interesting to see how AI inference and real-world capture are starting to converge.

All it takes is a single plane ✈️😉 Here's a super simple motion graphics hack to present your 3D scans - 0 skills required!

@Nagashi_000 Interesting idea! I will give it a try

@KIRI_Engine_App That's great. What happens if you use 360° video as video material? If you cut out the 360° video into frames, generate 3DGS for each frame, and then stitch together the generated 3DGS, does that mean you could create a 4DGS that better recreates the space?

One of our team members recently hacked together a lightweight local tool for dynamic 3DGS🤩 It takes a video, generates a 3DGS for each frame, and plays the result back as a timeline sequence for a pseudo-4DGS-style effect🤔 It’s still experimental, but the goal was to make this kind of workflow much easier to try locally. Would this be something you’d be interested in seeing more of?🥳

@DeKa95059471215 yes😊

@KIRI_Engine_App Mac?

@DonKikon_Art Thanks for the feedback! We'll figure out releases 😄

@KIRI_Engine_App Ohhhh yeahhh!!!! Do want!

@larecette404 Glad you like it! It’s still pretty experimental right now, but if there’s enough interest, we’d be happy to clean it up and put it on GitHub.

@KIRI_Engine_App I like it, will it be available on GitHub?

The 3D → plushie → 3D loop is complete. 1️⃣made a 3D model based on @0manaco0's 2️⃣turned the model into a pattern with @PlushifyLab 3️⃣tried my best to sew, resulting in a messed up plushie 4️⃣3D scanned the plushie with @KIRI_Engine_App 5️⃣rigged and set up the model for VTubing

@monamiralgames @0manaco0 @PlushifyLab This is so insanely cute! Love those floppy moves - reminds me a bit of human fall flat 🩷

This is a single Gaussian Splat scan, not a .PLY sequence or 4DGS. What would you call this, 3.5D? We trained a scan on various frames of an animation per view, resulting in a single file and no massive file sequence. The results may seem limited, but image this for a full werewolf transformation or other VFX?! What else should we try to "morph" using this method?

We removed the export limit from the our Free Plan🎉 It used to be 3 exports per week🤢🤮👎 Now it’s unlimited!!!!🥳 Free plan includes: • Unlimited Photo Scan & LiDAR scans • Unlimited exports • Photogrammetry with up to 150 photos per project • High-poly export with dynamic decimation Advanced Features like 3DGS→Mesh, AI PBR, and Quad-mesh retopo remain in Pro.