Bo Dai

29

Papers

1,295

Total Citations

Papers (29)

Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering

Generalized Predictive Model for Autonomous Driving

MotionLCM: Real-time Controllable Motion Generation via Latent Consistency Model

Unified Human-Scene Interaction via Prompted Chain-of-Contacts

LN3Diff: Scalable Latent Neural Fields Diffusion for Speedy 3D Generation

DiffMorpher: Unleashing the Capability of Diffusion Models for Image Morphing

Point Cloud Pre-training with Diffusion Models

Inference-Aware Fine-Tuning for Best-of-N Sampling in Large Language Models

ScaMo: Exploring the Scaling Law in Autoregressive Motion Generation Model

Efficient Online Reinforcement Learning for Diffusion Policy

PACER+: On-Demand Pedestrian Animation Controller in Driving Scenarios

DRiVE: Diffusion-based Rigging Empowers Generation of Versatile and Expressive Characters

Horizon-GS: Unified 3D Gaussian Splatting for Large-Scale Aerial-to-Ground Scenes

GAS: Generative Avatar Synthesis from a Single Image

Keyframe-Guided Creative Video Inpainting

BerfScene: Bev-conditioned Equivariant Radiance Fields for Infinite 3D Scene Generation

Multi-identity Human Image Animation with Structural Video Diffusion

ObjectGS: Object-aware Scene Reconstruction and Scene Understanding via Gaussian Splatting

GausSim: Foreseeing Reality by Gaussian Simulator for Elastic Objects

TokenHSI: Unified Synthesis of Physical Human-Scene Interactions through Task Tokenization

BBox-Adapter: Lightweight Adapting for Black-Box Large Language Models

Provable Representation with Efficient Planning for Partially Observable Reinforcement Learning

Target Networks and Over-parameterization Stabilize Off-policy Bootstrapping with Function Approximation

REINFORCE Converges to Optimal Policies with Any Learning Rate

Make-It-Vivid: Dressing Your Animatable Biped Cartoon Characters from Text

EpiDiff: Enhancing Multi-View Synthesis via Localized Epipolar-Constrained Diffusion

Cinematic Behavior Transfer via NeRF-based Differentiable Filming

FlashGS: Efficient 3D Gaussian Splatting for Large-scale and High-resolution Rendering

EdgeTAM: On-Device Track Anything Model