Merlin Nimier-David

📢 We introduce PPISP, a physically plausible camera module for Gaussian Splatting that reduces floaters under varying appearance and predicts camera corrections for novel views such that they closely match what a real camera would output. #CV #3DGS

Dr.Jit+Mitsuba just added support for fused neural networks, hash grids, and function freezing to eliminate tracing overheads. This speeds up optimization & realtime jobs and enables custom Instant NGP and neural material/radiosity/path guiding projects. What will you do with it?

Here's a gaussian splat of a macro shot I've been making. superspl.at/view?id=cf6ac7…

Gen3C, our generative video model with precise camera control is out! Start from any image and generate a video following the camera trajectory of your choice.

Our #Siggraph25 work found a simple, nearly one-line change that greatly eases neural field optimization for a wide variety of existing representations. “Stochastic Preconditioning for Neural Field Optimization” w/ @merlin_ND @_AlecJacobson @nmwsharp

🚀Excited to introduce GEN3C #CVPR2025, a generative video model with an explicit 3D cache for precise camera control. 🎥It applies to multiple use cases, including single-view and sparse-view NVS🖼️ and challenging settings like monocular dynamic NVS and driving simulation🚗. Project page: research.nvidia.com/labs/toronto-a…

@zhou_xian_ It would have been nice to acknowledge @taichi_lang in the announcement, especially since you showcase this large speedup and multi-platform support.

Nvidia brought GPU acceleration to robotic simulation, speeding up simulation speed by more than one order of magnitude compared to CPU-based simulation. This brought numerous amazing robotic skills to life by leveraging large-scale GPU-parallelized simulation. Genesis pushes up this speed by another order of magnitude. Note that the speed improvement is achieved with no compromise in simulation accuracy. 2/n

Everything you love about generative models — now powered by real physics! Announcing the Genesis project — after a 24-month large-scale research collaboration involving over 20 research labs — a generative physics engine able to generate 4D dynamical worlds powered by a physics simulation platform designed for general-purpose robotics and physical AI applications. Genesis's physics engine is developed in pure Python, while being 10-80x faster than existing GPU-accelerated stacks like Isaac Gym and MJX. It delivers a simulation speed ~430,000 faster than in real-time, and takes only 26 seconds to train a robotic locomotion policy transferrable to the real world on a single RTX4090 (see tutorial: genesis-world.readthedocs.io/en/latest/user…). The Genesis physics engine and simulation platform is fully open source at github.com/Genesis-Embodi…. We'll gradually roll out access to our generative framework in the near future. Genesis implements a unified simulation framework all from scratch, integrating a wide spectrum of state-of-the-art physics solvers, allowing simulation of the whole physical world in a virtual realm with the highest realism. We aim to build a universal data engine that leverages an upper-level generative framework to autonomously create physical worlds, together with various modes of data, including environments, camera motions, robotic task proposals, reward functions, robot policies, character motions, fully interactive 3D scenes, open-world articulated assets, and more, aiming towards fully automated data generation for robotics, physical AI and other applications. Open Source Code: github.com/Genesis-Embodi… Project webpage: genesis-embodied-ai.github.io Documentation: genesis-world.readthedocs.io 1/n

Following over 1.5 years of hard work (w/@njroussel& Rami Tabbara), we just released a brand-new version of Dr.Jit (v1.0), my lab's differentiable rendering compiler along with an updated Mitsuba (v3.6). The list of changes is insanely long—here is what we're most excited about🧵

It was great to be part of this project combining a diffusion model with differentiable physically based rendering!

@SheriefFYI I think it's an Apple Watch case, which explains the on-screen UI at least.

how is this not a shameless UI ripoff?

I am thankful to every individuals and companies which have allowed keeping Dear ImGui permissive and open source for many years. I do 100% intend to keep it as such. If your team relies on Dear ImGui consider reaching out to discuss ways of supporting the software.

📢Our new @NVIDIAAI paper is out! I'm happy to share Compact NGP with Learned Hash Probing, which offers NeRFs that are 3.5x smaller than Instant NGP at a ~1.26x training cost and equivalent (or faster) inference speeds! 🌐: nv-tlabs.github.io/compact-ngp/ 🎥: youtube.com/watch?v=3TEry8…

Proud to share that Adaptive Shells has been selected for a SIGGRAPH Asia Best Paper Award! #SIGGRAPHAsia2023 #NVIDIA

I'm really happy that I got to collaborate with such a talented team over the last few months, and proud of the result! Check out Adaptive Shells, which will be presented at SIGGRAPH Asia.

The free web version of "Physically Based Rendering: From Theory To Practice" is now based on the 4th edition of the book. Enjoy! (Link: pbr-book.org)

Did you know that inverse rendering can suffer from severe bias when the images are noisy (e.g. made using Monte Carlo methods). Our SIGGRAPH'23 paper dives into this overlooked issue (w/ @b_nicolet, Fabrice Rousselle, @_jannovak , Alexander Keller, and Thomas Müller)

@SebAaltonen cc @DelioVicini

Watching a recent differentiable SDF presentation (about neural mesh generation): rgl.epfl.ch/publications/V… Wondering if they could calculate the approximate ray surface distance using my triangulation algorithm instead... ShaderToy version: shadertoy.com/view/lsKcDD

Happy April Fools' day everyone! To celebrate the occasion, here is "Noodles", a small path tracer built entirely inside @Blender's shader node editor: tizianzeltner.com/projects/noodl… 🍜 Alternative title: a silly joke idea that got way out of hand. Co-conspirator: @b_nicolet

Look what arrived in the mail today!

@oservieres @Bouletcorp Et il faut aller voir son Patreon, c'est un niveau incroyable.

@Bouletcorp C'est très particulier, mais Ian Hubert est une énorme brute, et sa chaîne youtube est une pépite. Des tutos très très TRÈS courts qui ne rentrent pas dans les détails mais qui donnent toutes les grandes pistes. C'est a voir rien que pour le plaisir.