nv-tlabs

Official implementation of Kimodo, a kinematic motion diffusion model for high-quality human(oid) motion generation.

Project Lyra: Open Generative 3D World Models

Ray tracing and hybrid rasterization of Gaussian particles

ViPE: Video Pose Engine for Geometric 3D Perception

[CVPR 2025 Highlight] GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control

Lift, Splat, Shoot: Encoding Images from Arbitrary Camera Rigs by Implicitly Unprojecting to 3D (ECCV 2020)

[CVPR 2025 Oral & Best Paper Finalist] Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Models

Unifying 3D Mesh Generation with Language Models

Adversarial skill embeddings for training reusable controllers for physically simulated characters.

[NeurIPS 2026] PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion

Official implementation of ARDY: Autoregressive Diffusion with Hybrid Representation for Interactive Human Motion Generation (SIGGRAPH 2026).

[CVPR 2023 Highlight] Neural Kernel Surface Reconstruction

Gated-Shape CNN for Semantic Segmentation (ICCV 2019)

Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes (CVPR 2021 Oral)

Cosmos-Transfer1-DiffusionRenderer: High-quality video de-lighting and re-lighting based on Cosmos video diffusion framework

Latent Point Diffusion Models for 3D Shape Generation

[ICLR 2026] ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation

Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer (NeurIPS 2019)

Implementation of Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players

Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models

[CVPR 2024 Highlight] XCube: Large-Scale 3D Generative Modeling using Sparse Voxel Hierarchies

[NeurIPS 2024] SCube: Instant Large-Scale Scene Reconstruction using VoxSplats

Official implementation of Déjà View: Looping Transformers for Multi-View 3D Reconstruction

STEAL - Learning Semantic Boundaries from Noisy Annotations (CVPR 2019)

[ICCV 2025] PartField: Learning 3D Feature Fields for Part Segmentation and Beyond

Physically-Plausible Image Signal Processing (PPISP) for Radiance Field Reconstruction

[CVPR'25 Oral] Official implementation for "DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models"

Encoder-Free Unified Image and Video Understanding and Generation

NVIDIA OmniDreams is a world model that generates photorealistic video for autonomous-driving simulation in real time.

Official implementation for SIGGRAPH 2023 paper "Learning Physically Simulated Tennis Skills from Broadcast Videos"

Code for "ATISS: Autoregressive Transformers for Indoor Scene Synthesis", NeurIPS 2021

[CVPR'26] TokenGS: Decoupling 3D Gaussian Prediction from Pixels with Learnable Tokens

[NeurIPS 2024] L4GM: Large 4D Gaussian Reconstruction Model

Learning to Simulate Dynamic Environments with GameGAN (CVPR 2020)

Score-Based Generative Modeling with Critically-Damped Langevin Diffusion

Learning Deformable Tetrahedral Meshes for 3D Reconstruction (NeurIPS 2020)

Official repo for SemanticGAN https://nv-tlabs.github.io/semanticGAN/

Meta-Sim: Learning to Generate Synthetic Datasets (ICCV 2019)

Code for CVPR 2022 paper "Generating Useful Accident-Prone Driving Scenarios via a Learned Traffic Prior"

NVIDIA Fixer is a single-step diffusion model that removes reconstruction artifacts from rendered views for autonomous vehicle and robotics simulation.

Neural Harmonic Textures for High-Quality Primitive Based Neural Reconstruction

Grid-Free Monte Carlo Solvers for Physics Simulations Involving Partial Differential Equations

Implementation of "Multi-Track Timeline Control for Text-Driven 3D Human Motion Generation" from CVPR Workshop on Human Motion Generation 2024.

[ICCV 2025] InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models

[NeurIPS 2025] UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting

Official implementation of TRACE, the TRAjectory Diffusion Model for Controllable PEdestrians, from the CVPR 2023 paper: "Trace and Pace: Controllable Pedestrian Animation via Guided Trajectory...

[SIGGRAPH 2024] Diffusion Texture Painting

Official implementation of PACER, Pedestrian Animation ControllER, of CVPR 2023 paper: "Trace and Pace: Controllable Pedestrian Animation via Guided Trajectory Diffusion".

Differentially Private Diffusion Models

Code release for DriveGAN (CVPR 2021)

GENIE: Higher-Order Denoising Diffusion Solvers

Official implementation of TeSMo, a method for text-controlled scene-aware motion generation, from the ECCV 2024 paper: "Generating Human Interaction Motions in Scenes with Text Control".

project page: https://nv-tlabs.github.io/big-datasetgan/

VoMP: Predicting Volumetric Mechanical Property Fields

Context-Matched Distillation: Teacher Causality for Autoregressive Video Distillation

Learning to Evaluate Perception Models Using Planner-Centric Metrics

[ICML 2026] Official code for paper: Test-Time Training with KV Binding Is Secretly Linear Attention