Ruilong Li
@ruilong_li
Research Scientist @NVIDIAAI | Alumni of @berkeley_ai, @GoogleAI, @RealityLabs, @USC_ICT, @Tsinghua_Uni
Excited to announce 🚀gsplat v1.0🚀: a ⏩efficient⏩ CUDA backend for 3D Gaussian Splatting! docs.gsplat.studio A drop-in replacement of the official impl. with: - Up to 2x faster training; - Up to 4x less GPU memory; - Render millions of GSs in real-time; - And more;
Full episode dropping soon! Geeking out with @arthurallshire @redstone_hong on VideoMimic videomimic.net Co-hosted by @chris_j_paxton @micoolcho
Ping us if you would be interested in this summer internship and potentially extending to fall semester. It’s a great team with a long of strong minds! More about us: research.nvidia.com/labs/toronto-a…
📢📢We have a last-minute internship opening on my team at @NVIDIAAI for this summer. If you are interested and have experience with large feedforward reconstruction models or post-training image/video diffusion models, please get in touch!
Some problems can’t be rushed—they can only be done step by step, no matter how many people or processors you throw at them. We’ve scaled AI by making everything bigger and more parallel: Our models are parallel. Our scaling is parallel. Our GPUs are parallel. But what if the…
Trained directly on @insta360 X5 circular fisheyes with @NVIDIAAIDev 3DGUT, and rendered using a fisheye camera in the gsplat viewer. Princess of Wales Conservatory, Kew Gardens, London. #NVIDIA3DGUT #NVIDIASweepstakes #3DGS
Thrilled to see this work drop the same week as our “Camera as Relative Positional Encoding”: Both embrace the same spirit: ditch absolute references, model pairwise relative relations — and everything just works better. Cheers! Link to our paper🔗 liruilong.cn/prope/
🔥 π³ is here! 🔥 The new way to do 3D visual geometry. 🧠 Core Idea: A feed-forward net that doesn‘t care about input order. No more reference view! ✅ Permutation-Equivariant 🔄 ✅ Reference-Free 🆓 ✅ Highly Scalable 📈 ✅ SOTA Results 🚀
developer.nvidia.com/blog/cutlass-p… marks the start of a short series of blogposts about CUTLASS 3.x and CuTe that we've been meaning to write for years. There are a few more parts to come still, hope you enjoy!
Model and training code for LaCT on language model, AR video gen and novel view synthesis are released, also have a TTT layer implementation with sequence parallel supported. Both object-centric and scene-level view synthesis checkpoints are released 🤓— come play!
Bored of linear recurrent memories (e.g., linear attention) and want a scalable, nonlinear alternative? Our new paper “Test-Time Training Done Right” propose LaCT (Large Chunk Test-Time Training) — a highly efficient, massively scalable nonlinear memory with: 💡 Pure PyTorch…
wow decent!
Training directly on the circular fisheyes from 360 cameras like @insta360 X5 and @KandaoVR Qoocam 3 Ultra using @NVIDIAAIDev 3DGUT in gsplat 👀 John Trundle Highwalk, Barbican, London #NVIDIA3DGUT #NVIDIASweepstakes #gaussiansplatting #3DGS
This is awesome! An awesome project of using gsplat's API from C++ side! Code is very clean and nicely structured. Highly recommend checking out.
My C++ 3DGS implementation has transitioned to the gsplat backend and is now licensed under Apache 2.0. - Supports MCMC densification by default. - Includes a fused bilateral grid implementation. - A basic viewer, contributed by the community, is available with more features in…
It was a fun chat!
What do you ACTUALLY need to create 3D Gaussian Splats? Spoiler: Point clouds aren't required! Just joined @ruilong_li and @JaredHeinly to break down everything about Gaussian Splatting. Link to episode: share.transistor.fm/s/48d798f2 #ComputerVision #3D #GaussianSplatting #AI
🚀 We just open-sourced Cosmos DiffusionRenderer! This major upgrade brings significantly improved video de-lighting and re-lighting—powered by NVIDIA Cosmos and enhanced data curation. Released under Apache 2.0 and Open Model License. Try it out! 🔗 github.com/nv-tlabs/cosmo…
🚀 Introducing DiffusionRenderer, a neural rendering engine powered by video diffusion models. 🎥 Estimates high-quality geometry and materials from videos, synthesizes photorealistic light transport, enables relighting and material editing with realistic shadows and reflections
Excited to present VideoMimic this week at #CVPR2025! 🎥🤖 📌 POETs Workshop "Embodied Humans" Spotlight Talk | June 12, Thu, -10:10 | Room 101B 📌 Agents in Interaction: From Humans to Robots Poster #182-#201 | June 12, Thu, -12:15 | ExHall D Come by and chat!…
🚀 DiffusionRenderer is now open-source! Check out the code and model: github.com/nv-tlabs/diffu… We will present at #CVPR2025 this Sunday, June 15: 🗣️ Oral Session 6A: 1:00–2:30 PM CDT, Karl Dean Grand Ballroom 🖼️ Poster: 4:00–6:00 PM CDT, ExHall D (Poster #29)
🚀 Introducing DiffusionRenderer, a neural rendering engine powered by video diffusion models. 🎥 Estimates high-quality geometry and materials from videos, synthesizes photorealistic light transport, enables relighting and material editing with realistic shadows and reflections
🚀 Just in time for the #CVPR rush: we’ve released the code for Difix3D+ — a Best Paper Award candidate! 🔧 Try out the code & demos: github.com/nv-tlabs/Difix… 🎤 Oral (June 15): 1:00–1:15 PM CDT, Karl Dean Grand Ballroom 🖼️ Poster: 4:00–6:00 PM CDT, ExHall D (#57) Join us! @CVPR
🚀 Difix3D+ is now open-sourced! Check out the code and try the demo: github.com/nv-tlabs/Difix… We're presenting at #CVPR2025 this Sunday, June 15 — come say hi! 🗣️ Oral: 1:00–1:15 PM CDT, Karl Dean Grand Ballroom 🖼️ Poster: 4:00–6:00 PM CDT, ExHall D (Poster #57)
Excited to host @ZGojcic talk next week on @nvidia DiffusionRenderer, a new technique for neural rendering. It approximates how light behaves in the real world and can turn daytime scenes into night, sunny scenes into cloudy ones, and so on. It combines inverse and forward…
We are excited to share Cosmos-Drive-Dreams 🚀 A bold new synthetic data generation (SDG) pipeline powered by world foundation models—designed to synthesize rich, challenging driving scenarios at scale. Models, Code, Dataset, Tookit are released. Website:…
Curious about 3D Gaussians, simulation, rendering and the latest from #NVIDIA? Come to the NVIDIA Kaolin Library live-coding session at #CVPR2025, powered by a cloud GPU reserved especially for you. Wed, Jun 11, 8-noon. Bring your laptop! tinyurl.com/nv-kaolin-cvpr…
The time for new architectures is over? Not quite! SeNaTra, a native segmentation backbone, is waiting, let's see how it works 🧵arxiv.org/abs/2505.16993