Picture for Hao Su

Hao Su

Member, IEEE

When Should We Prefer State-to-Visual DAgger Over Visual Reinforcement Learning?

Add code
Dec 18, 2024
Viaarxiv icon

Policy Decorator: Model-Agnostic Online Refinement for Large Policy Model

Add code
Dec 18, 2024
Viaarxiv icon

ManiSkill-ViTac 2025: Challenge on Manipulation Skill Learning With Vision and Tactile Sensing

Add code
Nov 19, 2024
Viaarxiv icon

3D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generation

Add code
Oct 24, 2024
Figure 1 for 3D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generation
Figure 2 for 3D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generation
Figure 3 for 3D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generation
Figure 4 for 3D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generation
Viaarxiv icon

Reward-free World Models for Online Imitation Learning

Add code
Oct 17, 2024
Viaarxiv icon

ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI

Add code
Oct 01, 2024
Figure 1 for ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI
Figure 2 for ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI
Figure 3 for ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI
Figure 4 for ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI
Viaarxiv icon

ConDense: Consistent 2D/3D Pre-training for Dense and Sparse Features from Multi-View Images

Add code
Aug 30, 2024
Figure 1 for ConDense: Consistent 2D/3D Pre-training for Dense and Sparse Features from Multi-View Images
Figure 2 for ConDense: Consistent 2D/3D Pre-training for Dense and Sparse Features from Multi-View Images
Figure 3 for ConDense: Consistent 2D/3D Pre-training for Dense and Sparse Features from Multi-View Images
Figure 4 for ConDense: Consistent 2D/3D Pre-training for Dense and Sparse Features from Multi-View Images
Viaarxiv icon

SpaRP: Fast 3D Object Reconstruction and Pose Estimation from Sparse Views

Add code
Aug 19, 2024
Figure 1 for SpaRP: Fast 3D Object Reconstruction and Pose Estimation from Sparse Views
Figure 2 for SpaRP: Fast 3D Object Reconstruction and Pose Estimation from Sparse Views
Figure 3 for SpaRP: Fast 3D Object Reconstruction and Pose Estimation from Sparse Views
Figure 4 for SpaRP: Fast 3D Object Reconstruction and Pose Estimation from Sparse Views
Viaarxiv icon

MeshFormer: High-Quality Mesh Generation with 3D-Guided Reconstruction Model

Add code
Aug 19, 2024
Viaarxiv icon

Decoupling General and Personalized Knowledge in Federated Learning via Additive and Low-Rank Decomposition

Add code
Jun 28, 2024
Viaarxiv icon