This is so cool !

My Canada VISA application for CVPR'23 got approved after 19 months ...

My group at NVIDIA is looking for interns for next year! If you are excited about research in dynamic scene understanding, perception & reconstruction, camera pose estimation, or even climate modelling, apply using the link below or contact me directly! nvidia.wd5.myworkdayjobs.com/en-US/NVIDIAEx…

To ensure temporal consistency and a global understanding of depth ranges: 1. Employ full attention across snippet instead of factorized spatio-temporal attention used in typical video models. 2. Align depth across snippets, leveraging overlap of frames across multiple snippets

Introducing MegaSaM! 🎥 Accurate, fast, & robust structure + camera estimation from casual monocular videos of dynamic scenes! MegaSaM outputs camera parameters and consistent video depth, scaling to long videos with unconstrained camera paths and complex scene dynamics!

📢🚗✨ Excited to announce InfiniCube, our scalable generative model for dynamic 3D driving scene generation with high fidelity and controllability! InfiniCube generates very large-scale (300m×400m ~ 100,000m^2), dynamic 3D driving scenes given HD maps, 3D bounding boxes, and

Kudos to the girl for speaking up and EDUCATING the mit prof

Neat !

The idea of using the diffusion model to perform forward rendering from imperfect G-buffers is super interesting !

Neat ! This can potentially benefit many latent NeRF/3DGS attempts.

painful to watch this - it’s like adult kindergarten with insanely high stakes

Fifth year grad students to incoming ones at the prospective student visit day:

Obtaining high-quality data is really the key to low-level vision tasks, and at Nvidia, we do have tools (Omniverse & DriveSim) to empower this !

2020: "NVS? Why bother?" 2021: "Wait, it's about representation!" 2023: "Actually, representation for rendering!" 2025: "Bro, it’s not representation—it’s the data."