Liang Pan

35

Papers

127

Total Citations

Papers (35)

Digital Life Project: Autonomous 3D Characters with Social Intelligence

Multi-Space Alignments Towards Universal LiDAR Segmentation

MVPaint: Synchronized Multi-View Diffusion for Painting Anything 3D

AvatarGO: Zero-shot 4D Human-Object Interaction Generation and Animation

🎧MOSPA: Human Motion Generation Driven by Spatial Audio

SIMS: Simulating Stylized Human-Scene Interactions with Retrieval-Augmented Script Generation

GeneMAN: Generalizable Single-Image 3D Human Reconstruction from Multi-Source Human Data

VideoLucy: Deep Memory Backtracking for Long Video Understanding

NeurIPS 2025arXiv

A Plug-and-Play Physical Motion Restoration Approach for In-the-Wild High-Difficulty Motions

Unsupervised 3D Shape Completion Through GAN Inversion

Versatile Multi-Modal Pre-Training for Human-Centric Perception

Neural MoCon: Neural Motion Control for Physically Plausible Human Motion Capture

TCTrack: Temporal Contexts for Aerial Tracking

PTTR: Relational 3D Point Cloud Object Tracking With Transformer

LaserMix for Semi-Supervised LiDAR Semantic Segmentation

OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generation

Generative Diffusion Prior for Unified Image Restoration and Enhancement

CSG-Stump: A Learning Friendly CSG-Like Representation for Interpretable Shape Parsing

Robo3D: Towards Robust and Reliable 3D Perception against Corruptions

ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model

SHERF: Generalizable Human NeRF from a Single Image

HuMMan: Multi-modal 4D Human Dataset for Versatile Sensing and Modeling

LiMoE: Mixture of LiDAR Representation Learners from Automotive Scenes

3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion

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

GauUpdate: New Object Insertion in 3D Gaussian Fields with Consistent Global Illumination

Diff2I2P: Differentiable Image-to-Point Cloud Registration with Diffusion Prior

Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data and Metric Perspectives

MotionStreamer: Streaming Motion Generation via Diffusion-based Autoregressive Model in Causal Latent Space

Closely Interactive Human Reconstruction with Proxemics and Physics-Guided Adaption

Variational Relational Point Completion Network

Garment4D: Garment Reconstruction from Point Cloud Sequences

Balanced Chamfer Distance as a Comprehensive Metric for Point Cloud Completion

Segment Any Point Cloud Sequences by Distilling Vision Foundation Models

InsActor: Instruction-driven Physics-based Characters