cagri taskin

architect - artist - - @oddviz co-founder

[cagri taskin @ccagritaskin]

Used Claude Fable 5 to create an ambient predator-prey simulation. Flocking fish, hunting sharks, swaying kelp — and a cursor the shoals fear. Pure dark-ocean atmosphere.

https://t.co/9CeNXAHPus

[cagri taskin @ccagritaskin]

I built Image Grid Maker: a free Windows app that turns thousands of photos into one clean grid — always a perfect rectangle, with aspect-ratio cropping, custom output size and borders.

Made entirely with Claude Opus 4.8. Link in comments.

[cagri taskin @ccagritaskin]

https://t.co/HPBualbE38

[Antoine Guédon @antoine_guedon]

What if you could turn any number of photos (3, 8, 15, or even 60) into one clean 3D surface (pts & mesh) with Flow Matching?

Check out our new work, Surflo: Consistent 3D Surface Flow Model with Global State. 🧵
1/n

🔗https://t.co/lBcJRgpfdg

[cagri taskin @ccagritaskin]

So Claude Fable 5 and Mythos 5 were effectively shut off for users because the US government cited vague “national security” concerns?
Based on what Anthropic describes, there’s no clear public evidence, no transparent process, just a forced disruption.

https://t.co/guENf087j8

[cagri taskin @ccagritaskin]

@kotohibi_3d what a coincidence! I hit the bug and hotfix in minutes. ty for instructions.

[stevibe @stevibe]

Claude Opus 4.7 vs 4.8 side-by-side canvas test

[cagri taskin @ccagritaskin]

@kotohibi_3d I just bought 'Extract Sharpest Frame ( GUI, Windows binary edition ) Single user license' from 'kotohibi-cg'. https://t.co/EKdtMVYlOg #booth_pm

[cagri taskin @ccagritaskin]

Stork just pulled a Top Gun-tier corkscrew evasive maneuver!

[Kate from Kharkiv @BohuslavskaKate]

Some brain-dead russians tried to hit a live Ukrainian stork with a drone, but the bird completely outmaneuvered it.
Idiots 🤦‍♀️

[kotohibi @kotohibi_3d]

This slide is a workflow for mixed 360 and planar images for 3DGS by my tools. new version is coming soon. Not only drone but also handheld are available, of course. Let's add a twist for your 3DGS.😀
Google slide : https://t.co/TtW9JOynNy

[Joseph Azar @parazar]

Wild what's possible now: I can ship a 3D experience prototype to a museum in a single afternoon.
Everything coded and generated with Omma, leveraging AI to iterate fast.

3D models generated in Omma. Scene built on top of Three.js. Prompted for dissolving shaders + particle systems?

[Nathie @NathieVR]

This guy created a VR experience that literally lets you tear down reality.

[The Water Museum @WaterMuseum_]

They modeled everything flat and then turned it into a sphere, this is insane

[ア @yuruyurau]

a=(x,y,d=mag(k=4*cos(x/21),e=y/8-20))=>circle((q=3*sin(k*2)+.3/k+sin(y/19)*k*(9+2*sin(e*14-d*3+t*2)))+50*cos(c=d-t)+200,q*sin(c)+d*39-475,k*k>15?2:1)
t=0,draw=$=>{t||createCanvas(w=400,w);background(9).noStroke().fill(w,116);for(t+=PI/240,i=1e4;i--;)a(i,i/235)}#つぶやきProcessing

[Yakov @yak32]

I finally finished my Gaussian Splat based FPS demo.

It's a @playcanvas project, runs in a browser. On a real photoscan. With physics, baked lighting, pathfinding NPCs.

Here's how 👇

[Sadao Tokuyama @tokufxug]

3Dスキャンの常識が変わる。

最新の「Seen2Scene」は、不完全なスキャンから見えない領域をAIが自動補完し、高精細な3D空間を再構築。

実世界のデータから直接学習することで、複雑な配置も現実そのままに再現。

現実を丸ごとデジタル化する「世界モデル」の基盤がいよいよ整ってきた。

[Azad Balabanian @Azadux]

this might be the most detailed splat i've made yet - @lichtfeldstudio MRNF 115k steps

Church of the Holy Apostles - Ani, Turkey (2022) https://t.co/tdhSfLnw50

[cagri taskin @ccagritaskin]

@ChampagneMan @objktcom bellissimo choice. 🙏