Siyu Tang

50

Papers

173

Total Citations

Papers (50)

Optimizing Diffusion Noise Can Serve As Universal Motion Priors

Improving 2D Feature Representations by 3D-Aware Fine-Tuning

EgoGen: An Egocentric Synthetic Data Generator

Morphable Diffusion: 3D-Consistent Diffusion for Single-image Avatar Creation

EgoM2P: Egocentric Multimodal Multitask Pretraining

Degrees of Freedom Matter: Inferring Dynamics from Point Trajectories

DeGauss: Dynamic-Static Decomposition with Gaussian Splatting for Distractor-free 3D Reconstruction

Multi-View 3D Point Tracking

Multiple People Tracking by Lifted Multicut and Person Re-Identification

Generating Descriptions With Grounded and Co-Referenced People

Joint Graph Decomposition & Node Labeling: Problem, Algorithms, Applications

ArtTrack: Articulated Multi-Person Tracking in the Wild

Local Temporal Bilinear Pooling for Fine-Grained Action Parsing

Learning to Dress 3D People in Generative Clothing

Generating 3D People in Scenes Without People

LEAP: Learning Articulated Occupancy of People

Locally Aware Piecewise Transformation Fields for 3D Human Mesh Registration

We Are More Than Our Joints: Predicting How 3D Bodies Move

On Self-Contact and Human Pose

SCALE: Modeling Clothed Humans with a Surface Codec of Articulated Local Elements

COAP: Compositional Articulated Occupancy of People

Context-Aware Sequence Alignment Using 4D Skeletal Augmentation

The Wanderings of Odysseus in 3D Scenes

Human-Aware Object Placement for Visual Environment Reconstruction

Accurate 3D Body Shape Regression Using Metric and Semantic Attributes

HARP: Personalized Hand Reconstruction From a Monocular RGB Video

End-to-End Learning for Graph Decomposition

The Power of Points for Modeling Humans in Clothing

Learning Motion Priors for 4D Human Body Capture in 3D Scenes

3D Segmentation of Humans in Point Clouds with Synthetic Data

Guided Motion Diffusion for Controllable Human Motion Synthesis

Probabilistic Human Mesh Recovery in 3D Scenes from Egocentric Views

Dynamic Point Fields

Forecasting Human-Object Interaction: Joint Prediction of Motor Attention and Actions in First Person Video

ETH-XGaze: A Large Scale Dataset for Gaze Estimation under Extreme Head Pose and Gaze Variation

EgoBody: Human Body Shape and Motion of Interacting People from Head-Mounted Devices

SAGA: Stochastic Whole-Body Grasping with Contact

Compositional Human-Scene Interaction Synthesis with Semantic Control

KeypointNeRF: Generalizing Image-Based Volumetric Avatars Using Relative Spatial Encoding of Keypoints

ARAH: Animatable Volume Rendering of Articulated Human SDFs

Synthesizing Diverse Human Motions in 3D Indoor Scenes

VolumetricSMPL: A Neural Volumetric Body Model for Efficient Interactions, Contacts, and Collisions

UniPhys: Unified Planner and Controller with Diffusion for Flexible Physics-Based Character Control

RoHM: Robust Human Motion Reconstruction via Diffusion

3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting

IntrinsicAvatar: Physically Based Inverse Rendering of Dynamic Humans from Monocular Videos via Explicit Ray Tracing

Subgraph Decomposition for Multi-Target Tracking

DeepCut: Joint Subset Partition and Labeling for Multi Person Pose Estimation

MATE: Plugging in Model Awareness to Task Embedding for Meta Learning

MetaAvatar: Learning Animatable Clothed Human Models from Few Depth Images