Research

Explore our portfolio of projects spanning 3D vision, generative models, computational imaging, robotics, and more.

From Blurry to Believable: Enhancing Low-quality Talking Heads with 3D Generative Priors
From Blurry to Believable: Enhancing Low-quality Talking Heads with 3D Generative Priors
2025
Generative Models3D Vision
Gen2Act: Human Video Generation in Novel Scenarios enables Generalizable Robot Manipulation
Gen2Act: Human Video Generation in Novel Scenarios enables Generalizable Robot Manipulation
2025
RoboticsGenerative Models
SceneFactor: Factored Latent 3D Diffusion for Controllable 3D Scene Generation
SceneFactor: Factored Latent 3D Diffusion for Controllable 3D Scene Generation
2025
3D VisionGenerative Models
DressRecon: Freeform 4D Human Reconstruction from Monocular Video
DressRecon: Freeform 4D Human Reconstruction from Monocular Video
2025
3D Vision
Turbo3D: Ultra-fast Text-to-3D Generation
Turbo3D: Ultra-fast Text-to-3D Generation
2025
3D VisionGenerative Models
Generative Photomontage
Generative Photomontage
2025
Generative Models
GAS: Generative Avatar Synthesis from a Single Image
GAS: Generative Avatar Synthesis from a Single Image
2025
Generative Models3D Vision
DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion
DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion
2025
3D Vision
Hamba: Single-view 3D Hand Reconstruction with Graph-guided Bi-Scanning Mamba
Hamba: Single-view 3D Hand Reconstruction with Graph-guided Bi-Scanning Mamba
2025
3D VisionRecognition & Detection
Generating Physically Stable and Buildable Brick Structures from Text
Generating Physically Stable and Buildable Brick Structures from Text
2025
Generative Models3D Vision
Neural Inverse Rendering from Propagating Light
Neural Inverse Rendering from Propagating Light
2025
Computational Imaging
SplatSim: Zero-Shot Sim2Real Transfer of RGB Manipulation Policies
SplatSim: Zero-Shot Sim2Real Transfer of RGB Manipulation Policies
2025
Robotics3D Vision
Geometric Red-Teaming for Robotic Manipulation
Geometric Red-Teaming for Robotic Manipulation
2025
Robotics
AerialMegaDepth: Learning Aerial-Ground Reconstruction and View Synthesis
AerialMegaDepth: Learning Aerial-Ground Reconstruction and View Synthesis
2025
3D VisionScene Understanding
Unsupervised Identification of Protein Compositions via Content-Transformation Disentanglement
Unsupervised Identification of Protein Compositions via Content-Transformation Disentanglement
2025
Recognition & Detection
Spatially Varying Autofocus
Spatially Varying Autofocus
2025
Computational Imaging
Generating Multi-Image Synthetic Data for Text-to-Image Customization
Generating Multi-Image Synthetic Data for Text-to-Image Customization
2025
Generative Models
ArticuBot: Learning Universal Articulated Object Manipulation Policy
ArticuBot: Learning Universal Articulated Object Manipulation Policy
2025
Robotics
Unifying 2D and 3D Vision-Language Understanding
Unifying 2D and 3D Vision-Language Understanding
2025
Scene Understanding
Neural Eulerian Scene Flow Fields
Neural Eulerian Scene Flow Fields
2025
3D VisionScene Understanding
TAPIP3D: Tracking Any Point in Persistent 3D Geometry
TAPIP3D: Tracking Any Point in Persistent 3D Geometry
2025
3D Vision
DiffCAM: Data-Driven Saliency Maps by Capturing Feature Differences
DiffCAM: Data-Driven Saliency Maps by Capturing Feature Differences
2025
Recognition & Detection
Video Depth without Video Models
Video Depth without Video Models
2025
3D VisionScene Understanding
Taming 3D Gaussian Splatting for High-Quality Rendering
Taming 3D Gaussian Splatting for High-Quality Rendering
2025
3D Vision
Tactile DreamFusion: Exploiting Tactile Sensing for 3D Generation
Tactile DreamFusion: Exploiting Tactile Sensing for 3D Generation
2024
Robotics3D Vision
ODIN: A Single Model for 2D and 3D Perception
ODIN: A Single Model for 2D and 3D Perception
2024
Recognition & DetectionScene Understanding
Distilling Diffusion Models into Conditional GANs
Distilling Diffusion Models into Conditional GANs
2024
Generative Models
Visible to Thermal Light Transport
Visible to Thermal Light Transport
2024
Computational Imaging
Z-Splat: Z-axis Gaussian Splatting for Camera-Sonar Fusion
Z-Splat: Z-axis Gaussian Splatting for Camera-Sonar Fusion
2024
3D VisionComputational Imaging
Spectral Sub-Surface Scattering
Spectral Sub-Surface Scattering
2024
Computational Imaging
Coherence as Texture: Passive Textureless 3D Reconstruction by Self-Interference
Coherence as Texture: Passive Textureless 3D Reconstruction by Self-Interference
2024
Computational Imaging3D Vision
3D Diffuser Actor: Policy Diffusion with 3D Scene Representations
3D Diffuser Actor: Policy Diffusion with 3D Scene Representations
2024
Robotics3D Vision
DreamScene4D: Dynamic Multi-Object Scene Generation from Monocular Videos
DreamScene4D: Dynamic Multi-Object Scene Generation from Monocular Videos
2024
Generative Models3D Vision
3D reconstruction with fast dipole sums
3D reconstruction with fast dipole sums
2024
3D VisionComputational Imaging
Neural MP: A Generalist Neural Motion Planner
Neural MP: A Generalist Neural Motion Planner
2024
Robotics
Local Policies Enable Zero-shot Long-horizon Manipulation
Local Policies Enable Zero-shot Long-horizon Manipulation
2024
Robotics
HoloOcean: A Full-Features Marine Robotics Simulator
HoloOcean: A Full-Features Marine Robotics Simulator
2024
Robotics
HACMan++: Spatially-Grounded Motion Primitives for Manipulation
HACMan++: Spatially-Grounded Motion Primitives for Manipulation
2024
Robotics
Fabric Diffusion: High-Fidelity Texture Transfer for 3D Garments from In-The-Wild Images
Fabric Diffusion: High-Fidelity Texture Transfer for 3D Garments from In-The-Wild Images
2024
Generative Models3D Vision
Neural Feels with Neural Fields: Visuo-tactile Perception for In-hand Manipulation
Neural Feels with Neural Fields: Visuo-tactile Perception for In-hand Manipulation
2024
Robotics3D Vision
FlowBotHD: History-Aware Diffuser Handling Ambiguities in Articulated Objects
FlowBotHD: History-Aware Diffuser Handling Ambiguities in Articulated Objects
2024
Robotics
MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation
MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation
2024
3D VisionGenerative Models
NormalFlow: Fast Contact-based Object 6DoF Pose Tracking with Tactile Sensors
NormalFlow: Fast Contact-based Object 6DoF Pose Tracking with Tactile Sensors
2024
Robotics
SplaTAM: Splat, Track and Map 3D Gaussians for Dense RGB-D SLAM
SplaTAM: Splat, Track and Map 3D Gaussians for Dense RGB-D SLAM
2024
3D Vision
Extreme Parkour with Legged Robots
Extreme Parkour with Legged Robots
2024
Robotics
Shape from Heat Flow
Shape from Heat Flow
2024
Computational Imaging3D Vision
Customizing Text-to-Image Models with a Single Image Pair
Customizing Text-to-Image Models with a Single Image Pair
2024
Generative Models
Objects as volumes: A stochastic geometry view of opaque solids
Objects as volumes: A stochastic geometry view of opaque solids
2024
3D Vision
Cameras as Rays: Pose Estimation via Ray Diffusion
Cameras as Rays: Pose Estimation via Ray Diffusion
2024
3D Vision
Neural Kaleidoscopic Space Sculpting
Neural Kaleidoscopic Space Sculpting
2023
Computational Imaging3D Vision
Neural Implicit Surface Reconstruction using Imaging Sonar
Neural Implicit Surface Reconstruction using Imaging Sonar
2023
Computational Imaging3D Vision
Split-Lohmann Multifocal Displays for VR
Split-Lohmann Multifocal Displays for VR
2023
Computational Imaging
3D Scanning with Position-Sensitive Detectors
3D Scanning with Position-Sensitive Detectors
2023
Computational Imaging3D Vision
Ablating Concepts in Text-to-Image Diffusion Models
Ablating Concepts in Text-to-Image Diffusion Models
2023
Generative Models
Passive Micron-scale Time-of-Flight with Sunlight Interferometry
Passive Micron-scale Time-of-Flight with Sunlight Interferometry
2023
Computational Imaging
Controllable Visual-Tactile Synthesis
Controllable Visual-Tactile Synthesis
2023
RoboticsGenerative Models
ITI-GEN: Inclusive Text-to-Image Generation
ITI-GEN: Inclusive Text-to-Image Generation
2023
Generative Models
Bottom Up Top Down Detection Transformers for Language Grounding in Images and Point Clouds
Bottom Up Top Down Detection Transformers for Language Grounding in Images and Point Clouds
2022
Recognition & DetectionScene Understanding
Automatic Estimation of Self-Reported Pain by Trajectory Analysis
Automatic Estimation of Self-Reported Pain by Trajectory Analysis
2022
Recognition & Detection
Ego4D: Around the World in 3,000 Hours of Egocentric Video
Ego4D: Around the World in 3,000 Hours of Egocentric Video
2022
Recognition & DetectionScene Understanding
Sequential Voting with Relational Box Fields for Active Object Detection
Sequential Voting with Relational Box Fields for Active Object Detection
2022
Recognition & DetectionRobotics
Differentiable Raycasting for Self-supervised Occupancy Forecasting
Differentiable Raycasting for Self-supervised Occupancy Forecasting
2022
3D VisionRobotics
Human-to-Robot Imitation in the Wild
Human-to-Robot Imitation in the Wild
2022
Robotics
Compositional and Scalable Object SLAM
Compositional and Scalable Object SLAM
2021
3D VisionRobotics
Dynamics-Regulated Kinematic Policy for Egocentric Pose Estimation
Dynamics-Regulated Kinematic Policy for Egocentric Pose Estimation
2021
Recognition & Detection
Detecting Invisible People
Detecting Invisible People
2021
Recognition & Detection