patricio gonzalez vivo
banner
patricio.io
patricio gonzalez vivo
@patricio.io
2K followers 330 following 2.3K posts
Research Engineer. Author of https://thebookofshaders.com/, https://lygia.xyz/ and https://pixelspiritdeck.com Mas:@[email protected] Web: patriciogonzalezvivo.com
Posts Media Videos Starter Packs
Hehe thank you my friend! Have been a ride. So many lessons from our experiences back-in-the-days of OF
Very excited to finally share openly what I have been working on, Project Graph. Adobe's newest app we started with a handful of talented individuals last year. I'm so proud of the team and what we are building.
Big day tomorrow, we are showing a new product at MAX that we started with a hand full of talented individuals. Very proud of what we had accomplish in just a little more than a year.
Good eye. There a sota edge detection model on the mix + a face detection for setting the center of the rays
❤️ now I feel compelled to rewatch them
Just order mine! I'm so excited to get it. I'm looking for a place in the US to re-print pixelspiritdeck.com, I would love to talk about it @knaveofcups.com. Would it be possible?
PixelSpirit
pixelspiritdeck.com
Will need this analog layer to dry before another plotting
Pixel3DMM
Fine-tuned DINO ViT for per-pixel surface normal and uv-coordinate prediction
code: github.com/SimonGiebenh...
page: simongiebenhain.github.io/pixel3dmm/
CAP4D
Generate animatable 4D avatars from any image(s) + driving video.
code: github.com/felixtaubner...
page: felixtaubner.github.io/cap4d/
Check lygia.xyz. It's not a proper polyfill for a language-level feature, but for functions.
LYGIA Shader Library
GLSL/HLSL/WGSL/Metal/CUDA Shader library
lygia.xyz
LongSplat a robust unposed 3D Gaussian Splatting for Casual Long Videos
web linjohnss.github.io/longsplat/
code github.com/NVlabs/LongS...
Fascinating idea of using lens blur as a camera fingerprint. Some hope in the era of synthetic everything
url blur-fields.github.io
MapAnything, a simple, end-to-end trained transformer model that directly regresses the factored metric 3D geometry of a scene given various types of inputs (images, calibration, poses, or depth).
code: github.com/facebookrese...
web: map-anything.github.io
Cleaver image encoding using gaussian splats

paper: arxiv.org/pdf/2407.01866
code: github.com/NYU-ICL/imag...
Yes that’s the plan. But first we are waiting for all the tariffs and imports to settle
Hi community, I have been cooking a new product at Adobe… fingers crossed 🤞 we would be hable to show a bit of it in this coming Adobe MAX. Send good vibes
Thank you for the replays and tips. TLDR we are living a transitions from search engines SEO to Generative Engines GEO. AI crawlers come to your site and train on it so it can give people answers; the “solution” is to let in those that can site you back, and put timers that feed at a health pase