Shubham Tulsiani

10

Papers

142

Total Citations

Papers (10)

Track2Act: Predicting Point Tracks from Internet Videos enables Generalizable Robot Manipulation

MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation

AerialMegaDepth: Learning Aerial-Ground Reconstruction and View Synthesis

UpFusion: Novel View Diffusion from Unposed Sparse View Observations

Turbo3D: Ultra-fast Text-to-3D Generation

UniPhy: Learning a Unified Constitutive Model for Inverse Physics Simulation

DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion

LightSwitch: Multi-view Relighting with Material-guided Diffusion

SceneFactor: Factored Latent 3D Diffusion for Controllable 3D Scene Generation

G-HOP: Generative Hand-Object Prior for Interaction Reconstruction and Grasp Synthesis