Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

i built a real-time 3d positional tracker in the browser renders in ~8-10ms

35,343 Aufrufe • vor 2 Tagen •via X (Twitter)

34 Kommentare

Profilbild von John Lindquist
John Lindquistvor 1 Tag

i did this 17 years ago 😉

Profilbild von vogel
vogelvor 1 Tag

nice!

Profilbild von John Lindquist
John Lindquistvor 1 Tag

i miss flash so much. imagine what astra could do with flash

Profilbild von vogel
vogelvor 2 Tagen

try it out here:

Profilbild von maria
mariavor 2 Tagen

ryan rouch, vogel browne

Profilbild von Salami
Salamivor 1 Tag

Stop with the indian ai linkden slop posts please there are multiple projects that already do this

Profilbild von Keegan
Keeganvor 1 Tag

Gotta pump those response numbers up

Profilbild von Caleb
Calebvor 2 Tagen

you should drop this as a space on 🤗

Profilbild von vogel
vogelvor 2 Tagen

how

Profilbild von Caleb
Calebvor 1 Tag

you can just ask your agent to deploy here, but saw the link above which is neat

Profilbild von vogel
vogelvor 1 Tag

nice:

Profilbild von 🔥 nor
🔥 norvor 1 Tag

which lab has the model "i" ?

Profilbild von Vladimir Cazacu | 𝕧𝕠𝕝𝕠𝕕▿
Vladimir Cazacu | 𝕧𝕠𝕝𝕠𝕕▿vor 1 Tag

That's insanely slow. I was doing fiducial marker tracking via webcam at 2-3ms in 2007.

Profilbild von Dak
Dakvor 1 Tag

so cool haha

Profilbild von Arthur Katcher
Arthur Katchervor 1 Tag

really cool

Profilbild von UNTITLED
UNTITLEDvor 1 Tag

What if you put all frames into a buffer (seperrate mode), do the tracking and render per frame and then play those frames? It will be slightly delayed, but accuracy should be very high right?

Profilbild von R
Rvor 1 Tag

ryanvoxel hehe

Profilbild von Gonza Pini
Gonza Pinivor 1 Tag

Welcome back Invizimals

Profilbild von bytesworms
byteswormsvor 1 Tag

so smooth 🔥

Profilbild von Kairo Salomo Chee Chin
Kairo Salomo Chee Chinvor 1 Tag

Interesting, i remember using some kind of synonym software to track my arm movements. completely forgot what it was.

Profilbild von Jonas
Jonasvor 1 Tag

How are you handling camera calibration? Normally for AprilTag tracking you take a bunch of photos of a calibration board at different positions and angles to calculate your camera intrinsics. If I were to guess, I assume you're just picking a reasonable assumption? That'll technically work, but AprilTag can actually have centimeter localization accuracy with good calibration, so it might be worth exploring using mrcal to generate good camera intrinsics.

Profilbild von Jason Ziegler
Jason Zieglervor 1 Tag

Neat!

Profilbild von KiLee
KiLeevor 1 Tag

Open source ?

Profilbild von Flat 18 - Web Design & Development
Flat 18 - Web Design & Developmentvor 1 Tag

That is a remarkably tight render loop for a browser-based tracker. The interesting product question is how you keep the interaction feeling stable when the input gets noisy.

Profilbild von neamtu
neamtuvor 1 Tag

No Dax Demon on it? Wtf is this demo

Profilbild von Rohan Mukherjee
Rohan Mukherjeevor 1 Tag

pretty cool!

Profilbild von Sindre Reino Trosterud
Sindre Reino Trosterudvor 1 Tag

Now make it track a mosquito

Profilbild von Sm17
Sm17vor 1 Tag

Man taking credit for clankers work!

Profilbild von Martin Solev
Martin Solevvor 1 Tag

But what's the use case? Very impressive btw!

Profilbild von Theo Ephraim 🧙‍♂️
Theo Ephraim 🧙‍♂️vor 1 Tag

I remember when doing any kind of decent CV in the browser was absolutely black magic. I even wrote a paper about realtime face detection using flash in 2009 😂

Profilbild von Van Kram
Van Kramvor 1 Tag

Try doing it without the robo-QR codes

Profilbild von Taggart Maher
Taggart Mahervor 1 Tag

not fast enough man

Profilbild von MixRoute AI
MixRoute AIvor 1 Tag

What is it called?

Profilbild von Prince Kushwaha
Prince Kushwahavor 1 Tag

Sub-10ms render times in the browser means you're hitting 100+ FPS locally—WebGPU and Wasm have made the browser a legitimate spatial computing engine.

Ähnliche Videos