Research Scientist at @GoogleDeepMind. UC Berkeley PhD 2020 + Duke 2014. 3D computer vision + graphics (NeRF!)
San Francisco, CA
Joined August 2011
- Tweets79
- Following149
- Followers1.7K
- Likes586
Pratul Srinivasan retweeted
Check out this work led by my student Brandon Zhao (@joiedeposter), with @pratul, @dorverbin, @jon_barron & Y. Wang! We use trainable pyramids to solve PDEs more accurately. I even got to learn a little about PDEs along the way - handy now that they seem to be all over the news!
Pratul Srinivasan retweeted
Introducing beignet, a new neural field architecture!
beignet uses a multiresolution fourier frequency pyramid to achieve faster convergence and higher precision when solving PDEs and finding singularities.
Teaser: here is beignet at work fitting an image (super fast)!
Pratul Srinivasan retweeted
Editing texture in 3DGS results in artifacts that become especially pronounced when the geometry changes.
Introducing Point-GT, which learns a UV map that adapts automatically to deformations with PAPR.
Joint w/ @yszhang170, @hekoshii and @_pratul_.
Project Website: zvict.github.io/pointgt/
Paper: arxiv.org/abs/2609.03341
Code: github.com/zvict/pointgt
Video: youtube.com/watch?v=aHKiGW_o…
If you are at #ECCV2026, come check out poster #206 this Thursday morning at 10:30am!
Pratul Srinivasan retweeted
Check out our work on 3D scene relighting! We rendered purely synthetic data and used it to fine-tune a video model for relighting. Even though it was only trained on synthetic data, it generalizes to real captured scenes—shadows, reflections, and all!
[1/6] We introduce GR3EN, a generative approach for relighting 3D environments. It gives you full control over every light source in a 3D scene — enabling adding new light sources, changing the color of existing ones or turning them off entirely. Shadows, reflections, and interreflections all update realistically.
Accepted to SIGGRAPH 2026 🔥
Project_page: gr3en-relight.github.io/
From: @GoogleDeepMind @GoogleResearch @UvA_Amsterdam
Pratul Srinivasan retweeted
[1/6] We introduce GR3EN, a generative approach for relighting 3D environments. It gives you full control over every light source in a 3D scene — enabling adding new light sources, changing the color of existing ones or turning them off entirely. Shadows, reflections, and interreflections all update realistically.
Accepted to SIGGRAPH 2026 🔥
Project_page: gr3en-relight.github.io/
From: @GoogleDeepMind @GoogleResearch @UvA_Amsterdam
Pratul Srinivasan retweeted
🎥 What if 3D capture could gracefully handle moving scenes and varying illumination?
🎯Come see how video models generate exactly the data you need at our poster, SimVS!
📍CVPR, June 14th (afternoon), Poster #60.
Pratul Srinivasan retweeted
⚡️ Introducing Bolt3D ⚡️
Bolt3D generates interactive 3D scenes in less than 7 seconds on a single GPU from one or more images.
It features a latent diffusion model that *directly* generates 3D Gaussians of seen and unseen regions, without any test time optimization.
🧵👇 (1/9)
Pratul Srinivasan retweeted
🚀 Introducing SimVS: our new method that simplifies 3D capture!
🎯 3D reconstruction assumes consistency—no dynamics or lighting changes—but reality constantly breaks this assumption.
✨ SimVS takes a set of inconsistent images and makes them consistent with a chosen frame.
Pratul Srinivasan retweeted
We’ll be presenting NeRF-Casting at SIGGRAPH Asia next week! NeRF-Casting enables photorealistic rendering of scenes with highly reflective surfaces—something that was previously impossible with models like Zip-NeRF and 3DGS. (1/6)
Pratul Srinivasan retweeted
(1/N) Flash Cache: Reducing Bias in Radiance Cache Based Inverse Rendering
Website: benattal.github.io/flash-cac…
tl;dr our #ECCV2024 (oral ✨) paper presents a new system for inverse rendering that is more physically accurate, and therefore less biased, than existing approaches.
Pratul Srinivasan retweeted
I'm going to present our work at the oral session tomorrow (Wednesday), 9am at #CVPR2024. Come check it out and hang out at the poster session
(ours is number 399) immediately after!
Introducing Eclipse, a method for recovering lighting and materials even from diffuse objects!
The key idea is that standard "NeRF-like" data has all we need: a photographer moving around a scene to capture it causes "accidental" lighting variations. dorverbin.github.io/eclipse/ (1/3)
Pratul Srinivasan retweeted
IllumiNeRF lets you relight objects in 3D. Instead of the classical inverse rendering approach — disentangling the object geometry, materials, and lighting — we use a relighting diffusion model to relight each input image and distill the relit samples into 3D by optimizing a latent NeRF.
This project was led by the talented @xmzhao_ who is completing his PhD and is currently on the job market!
Wondering how to easily relight an object?
Inverse rendering, maybe the first thing that comes to mind, is brittle and expensive due to differentiable Monte Carlo rendering.
Check out IllumiNeRF for simple, effective 3D relighting without it!
illuminerf.github.io
(1/n)
Pratul Srinivasan retweeted
Wondering how to easily relight an object?
Inverse rendering, maybe the first thing that comes to mind, is brittle and expensive due to differentiable Monte Carlo rendering.
Check out IllumiNeRF for simple, effective 3D relighting without it!
illuminerf.github.io
(1/n)
Pratul Srinivasan retweeted
IllumiNeRF enables relighting without expensive inverse rendering. We use a diffusion model trained to relight a single image, and turn its samples into a consistent 3D relit NeRF.
With @xmzhao_ (currently on the job market!) @_pratul_ @KeunhongP @rmbrualla @philipphenzler
Wondering how to easily relight an object?
Inverse rendering, maybe the first thing that comes to mind, is brittle and expensive due to differentiable Monte Carlo rendering.
Check out IllumiNeRF for simple, effective 3D relighting without it!
illuminerf.github.io
(1/n)
Pratul Srinivasan retweeted
IllumiNeRF
3D Relighting without Inverse Rendering
Existing methods for relightable view synthesis -- using a set of images of an object under unknown lighting to recover a 3D representation that can be rendered from novel viewpoints under a target illumination
Pratul Srinivasan retweeted
Check out CAT3D!
Image(s)-to-3D in 1 minute!
cat3d.github.io
Given any number of real or generated images, CAT3D uses a multi-view diffusion prior to create consistent novel views. These views are used to reconstruct a 3D scene using NeRF/3DGS.
Very excited to get this out. "CAT3D: Create Anything in 3D with Multi-View Diffusion Models"
Text->3D, image->3D, and few-view->3D, all in one package. SOTA few-view results, beautiful text results, trains in 1 minute, and renders at 60fps in a browser. cat3d.github.io/
The cat’s out of the bag!
Image(s) to 3D in one minute with a multi-view diffusion model.
cat3d.github.io
🌟 Create anything in 3D! 🌟 Introducing CAT3D: a new method that generates high-fidelity 3D scenes from any number of real or generated images in one minute, powered by multi-view diffusion models.
w/ lovely coauthors @holynski_, @poolio and an amazing team!
Pratul Srinivasan retweeted
Videos are cool and all...but everything's more fun when it's interactive.
Check out our new project, ✨CAT3D✨, that turns anything (text, image, & more) into interactive 3D scenes!
Don't miss the demo!!
cat3d.github.io/
🌟 Create anything in 3D! 🌟 Introducing CAT3D: a new method that generates high-fidelity 3D scenes from any number of real or generated images in one minute, powered by multi-view diffusion models.
w/ lovely coauthors @holynski_, @poolio and an amazing team!
Pratul Srinivasan retweeted
🌟 Create anything in 3D! 🌟 Introducing CAT3D: a new method that generates high-fidelity 3D scenes from any number of real or generated images in one minute, powered by multi-view diffusion models.
w/ lovely coauthors @holynski_, @poolio and an amazing team!
Pratul Srinivasan retweeted
Scientists, led by a team at Caltech, used AI and telescope data to create the first 3D video of mysterious bright flares around the supermassive black hole at the center of our galaxy.
caltech.edu/about/news/ai-an…
