Andrea Tagliasacchi

🔈🔈🔈 Check out our new paper!
Lead by Shakiba during her Google Deepmind internship.

RadFoam source code has arrived! (Apache-v2)
github.com/theialab/rad...

Belated happy Valentine's day 🥰

Hello everyone!! 👋
Excited to be here and share our latest work to get started!

RoMo: Robust Motion Segmentation Improves Structure from Motion

romosfm.github.io

Boost the performance of your SfM pipeline on dynamic scenes! 🚀 RoMo masks dynamic objects in a video, in a zero-shot manner.

Out of 🥰

Thank @martin-r-oswald.bsky.social for convincing me to cross-post 😜

Cross-post from x.com/taiyasaki/st...

And how does it perform? Very competitively, but the biggest gain is how much simpler everything becomes (after some gruesome CUDA coding 😋)

And how do you render/train this?
– sorting tricks? Nope.
– rely on Ray tracing libraries like Optix? No way!!

Just "walk" along the Voronoi cells! (you can think this as a generalization of a Digital Differential Analyzer, or DDA, to convex polygonal domains)

The core idea is to use a 3D polygonal mesh to perform differentiable optimization... but how to differentiate wrt. its combinatorial structure? Turns out the Voronoi diagram is differentiable w.r.t. the positions of its sites (while Delaunay triangulations are not w.r.t. edge flips)

"𝐑𝐚𝐝𝐢𝐚𝐧𝐭 𝐅𝐨𝐚𝐦: Real-Time Differentiable Ray Tracing"

A mesh-based 3D represention for training radiance fields from collections of images.

radfoam.github.io
arxiv.org/abs/2502.01157

Project co-lead by my PhD students Shrisudhan Govindarajan and Daniel Rebain, and w/ co-advisor Kwang Moo Yi

I meant platform adoption…
A true shift in users.

But indeed Twitter is being enshitified with more and more ads and less and less features.

I’ve tried to quit Facebook for g+, WhatsApp for Signal, and Twitter for… was it threads? There was another one.

I want to see adoption curves :)