Skip to content

@nv-tlabs NVIDIA Spatial Intelligence Lab (SIL)

All

    Repositories list

    121 repositories

    • 3dgrut

      Public
      Ray tracing and hybrid rasterization of Gaussian particles
      computer-visionray-tracing3d-reconstruction
      computer-visionray-tracing3d-reconstructiongaussian-splatting
      Python
      •
      Apache License 2.0
      •281•2.4k•59•7•Updated Sep 17, 2026Sep 17, 2026

    • cmd

      Public
      Context-Matched Distillation: Teacher Causality for Autoregressive Video Distillation
      Python
      •
      Other
      •2•62•0•1•Updated Sep 17, 2026Sep 17, 2026

    • vipe

      Public
      ViPE: Video Pose Engine for Geometric 3D Perception
      cameraslam3d
      cameraslam3ddepth-estimation
      Python
      •
      Other
      •175•2.1k•14•8•Updated Sep 10, 2026Sep 10, 2026

    • 3DGRUT-ArtiFixer

      Public
      Python
      •
      Apache License 2.0
      •7•17•1•1•Updated Sep 2, 2026Sep 2, 2026

    • wosx

      Public
      Grid-Free Monte Carlo Solvers for Physics Simulations Involving Partial Differential Equations
      C++
      •
      Apache License 2.0
      •11•147•0•0•Updated Sep 1, 2026Sep 1, 2026

    • learning-convex-decomposition

      Public
      Python
      •
      Other
      •3•70•0•0•Updated Aug 28, 2026Aug 28, 2026

    • NKSR

      Public
      [CVPR 2023 Highlight] Neural Kernel Surface Reconstruction
      computer-visiongraphicspoint-cloud
      computer-visiongraphicspoint-cloud3d-reconstructionneural-kernel
      Cuda
      •
      Other
      •72•995•37•0•Updated Aug 25, 2026Aug 25, 2026

    • sil-wheel

      Public
      SIL-Wheel is a one-stop shop for the data-related tooling for working on large-scale video model training.
      data-visualizationvideo-processingdata-analysis
      data-visualizationvideo-processingdata-analysisdata-curation
      Python
      •
      Apache License 2.0
      •1•15•0•1•Updated Aug 12, 2026Aug 12, 2026

    • TokenGS

      Public
      [CVPR'26] TokenGS: Decoupling 3D Gaussian Prediction from Pixels with Learnable Tokens
      3d-reconstructionnovel-view-synthesis
      3d-reconstructionnovel-view-synthesis
      Python
      •
      Apache License 2.0
      •15•271•1•0•Updated Aug 5, 2026Aug 5, 2026

    • Fixer

      Public
      NVIDIA Fixer is a single-step diffusion model that removes reconstruction artifacts from rendered views for autonomous vehicle and robotics simulation.
      Python
      •
      Apache License 2.0
      •13•150•5•0•Updated Jul 30, 2026Jul 30, 2026

    • neural-harmonic-textures

      Public
      Neural Harmonic Textures for High-Quality Primitive Based Neural Reconstruction
      Python
      •
      Apache License 2.0
      •10•145•3•0•Updated Jul 30, 2026Jul 30, 2026

    • Project-Lyra

      Public
      HTML
      •0•1•0•0•Updated Jul 25, 2026Jul 25, 2026

    • omni-dreams

      Public
      NVIDIA Cosmos-Dreams (fka NVIDIA OmniDreams) is a world model that generates photorealistic video for autonomous-driving simulation in real time.
      Python
      •
      Apache License 2.0
      •22•338•3•1•Updated Jul 24, 2026Jul 24, 2026

    • ArtiFixer

      Public
      Python
      •
      Apache License 2.0
      •57•662•3•0•Updated Jul 22, 2026Jul 22, 2026

    • PiD

      Public
      PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion
      pixel-diffusiondiffusion-decoder
      pixel-diffusiondiffusion-decoder
      Python
      •
      Other
      •61•1.1k•11•0•Updated Jul 22, 2026Jul 22, 2026

    • lyra

      Public
      Project Lyra: Open Generative 3D World Models
      video3ddiffusion
      video3ddiffusiongaussiansworld-model
      Python
      •
      Apache License 2.0
      •233•2.3k•21•4•Updated Jul 20, 2026Jul 20, 2026

    • VoMP

      Public
      VoMP: Predicting Volumetric Mechanical Property Fields
      Python
      •
      Apache License 2.0
      •7•70•0•1•Updated Jul 17, 2026Jul 17, 2026

    • cascade-devkit

      Public
      DevKit for CASCADE: Causal Spatio-Temporal Analysis of Driving Environments
      aiautonomousdrivingcausal-reasoning
      aiautonomousdrivingcausal-reasoning
      Jupyter Notebook
      •
      Apache License 2.0
      •0•1•1•0•Updated Jul 16, 2026Jul 16, 2026

    • kimodo

      Public
      Official implementation of Kimodo, a kinematic motion diffusion model for high-quality human(oid) motion generation.
      Python
      •
      Apache License 2.0
      •396•3.6k•9•9•Updated Jul 13, 2026Jul 13, 2026

    • ardy

      Public
      Official implementation of ARDY: Autoregressive Diffusion with Hybrid Representation for Interactive Human Motion Generation (SIGGRAPH 2026).
      Python
      •
      Apache License 2.0
      •116•924•6•7•Updated Jul 10, 2026Jul 10, 2026

    • kimodo-viser

      Public
      Python
      •
      Apache License 2.0
      •15•23•0•1•Updated Jul 6, 2026Jul 6, 2026

    • XCube

      Public
      [CVPR 2024 Highlight] XCube: Large-Scale 3D Generative Modeling using Sparse Voxel Hierarchies
      computer-visiongraphicsvoxel
      computer-visiongraphicsvoxel3d-generationgenerative-ai
      Python
      •
      Other
      •42•549•20•1•Updated Jun 18, 2026Jun 18, 2026

    • Gamma-World

      Public
      Implementation of Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players
      roboticsvideo-gamemulti-agent
      roboticsvideo-gamemulti-agentaigcworldmodel
      Python
      •
      Apache License 2.0
      •14•663•1•0•Updated Jun 17, 2026Jun 17, 2026

    • GEN3C

      Public
      [CVPR 2025 Highlight] GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control
      camera-control3d-graphicsvideo-diffusion-model
      camera-control3d-graphicsvideo-diffusion-model
      Jupyter Notebook
      •
      Apache License 2.0
      •81•1.4k•32•1•Updated Jun 15, 2026Jun 15, 2026

    • ppisp

      Public
      Physically-Plausible Image Signal Processing (PPISP) for Radiance Field Reconstruction
      Python
      •
      Apache License 2.0
      •43•457•2•0•Updated Jun 11, 2026Jun 11, 2026

    • PartField

      Public
      [ICCV 2025] PartField: Learning 3D Feature Fields for Part Segmentation and Beyond
      open-world3dfeature-learning
      open-world3dfeature-learningpart-segmentation
      Python
      •
      Other
      •43•462•12•0•Updated Jun 2, 2026Jun 2, 2026

    • dvlt

      Public
      Official implementation of Déjà View: Looping Transformers for Multi-View 3D Reconstruction
      Python
      •
      Other
      •22•441•4•0•Updated Jun 1, 2026Jun 1, 2026

    • LuxDiT

      Public
      [NeurIPS 2025] LuxDiT: Lighting Estimation with Video Diffusion Transformer
      Python
      •
      Other
      •5•43•1•0•Updated May 1, 2026May 1, 2026

    • tttla

      Public
      [ICML 2026] Official code for paper: Test-Time Training with KV Binding Is Secretly Linear Attention
      0•50•0•0•Updated Apr 30, 2026Apr 30, 2026

    • UniRelight

      Public
      [NeurIPS 2025] UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting
      Jupyter Notebook
      •
      Other
      •6•131•1•0•Updated Apr 8, 2026Apr 8, 2026
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.