Picture for Sanja Fidler

Sanja Fidler

NVIDIA, University of Toronto, Vector Institute

Cosmos-Transfer1: Conditional World Generation with Adaptive Multimodal Control

Add code
Mar 18, 2025
Viaarxiv icon

GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control

Add code
Mar 05, 2025
Viaarxiv icon

Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Models

Add code
Mar 03, 2025
Viaarxiv icon

DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models

Add code
Jan 30, 2025
Figure 1 for DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models
Figure 2 for DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models
Figure 3 for DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models
Figure 4 for DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models
Viaarxiv icon

Cosmos World Foundation Model Platform for Physical AI

Add code
Jan 07, 2025
Figure 1 for Cosmos World Foundation Model Platform for Physical AI
Figure 2 for Cosmos World Foundation Model Platform for Physical AI
Figure 3 for Cosmos World Foundation Model Platform for Physical AI
Figure 4 for Cosmos World Foundation Model Platform for Physical AI
Viaarxiv icon

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

Add code
Dec 05, 2024
Figure 1 for InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models
Figure 2 for InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models
Figure 3 for InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models
Figure 4 for InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models
Viaarxiv icon

Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos

Add code
Dec 04, 2024
Figure 1 for Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos
Figure 2 for Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos
Figure 3 for Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos
Figure 4 for Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos
Viaarxiv icon

LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models

Add code
Nov 14, 2024
Figure 1 for LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models
Figure 2 for LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models
Figure 3 for LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models
Figure 4 for LLaMA-Mesh: Unifying 3D Mesh Generation with Language Models
Viaarxiv icon

ReMatching Dynamic Reconstruction Flow

Add code
Nov 01, 2024
Figure 1 for ReMatching Dynamic Reconstruction Flow
Figure 2 for ReMatching Dynamic Reconstruction Flow
Figure 3 for ReMatching Dynamic Reconstruction Flow
Figure 4 for ReMatching Dynamic Reconstruction Flow
Viaarxiv icon

SCube: Instant Large-Scale Scene Reconstruction using VoxSplats

Add code
Oct 26, 2024
Figure 1 for SCube: Instant Large-Scale Scene Reconstruction using VoxSplats
Figure 2 for SCube: Instant Large-Scale Scene Reconstruction using VoxSplats
Figure 3 for SCube: Instant Large-Scale Scene Reconstruction using VoxSplats
Figure 4 for SCube: Instant Large-Scale Scene Reconstruction using VoxSplats
Viaarxiv icon