Picture for Haiyang Liu

Haiyang Liu

Video Motion Graphs

Add code
Mar 26, 2025
Viaarxiv icon

Free-viewpoint Human Animation with Pose-correlated Reference Selection

Add code
Dec 23, 2024
Viaarxiv icon

Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition

Add code
Dec 19, 2024
Figure 1 for Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition
Figure 2 for Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition
Figure 3 for Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition
Figure 4 for Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition
Viaarxiv icon

GWQ: Gradient-Aware Weight Quantization for Large Language Models

Add code
Oct 30, 2024
Figure 1 for GWQ: Gradient-Aware Weight Quantization for Large Language Models
Figure 2 for GWQ: Gradient-Aware Weight Quantization for Large Language Models
Figure 3 for GWQ: Gradient-Aware Weight Quantization for Large Language Models
Figure 4 for GWQ: Gradient-Aware Weight Quantization for Large Language Models
Viaarxiv icon

TANGO: Co-Speech Gesture Video Reenactment with Hierarchical Audio Motion Embedding and Diffusion Interpolation

Add code
Oct 05, 2024
Viaarxiv icon

Global-Aware Enhanced Spatial-Temporal Graph Recurrent Networks: A New Framework For Traffic Flow Prediction

Add code
Jan 07, 2024
Figure 1 for Global-Aware Enhanced Spatial-Temporal Graph Recurrent Networks: A New Framework For Traffic Flow Prediction
Figure 2 for Global-Aware Enhanced Spatial-Temporal Graph Recurrent Networks: A New Framework For Traffic Flow Prediction
Figure 3 for Global-Aware Enhanced Spatial-Temporal Graph Recurrent Networks: A New Framework For Traffic Flow Prediction
Figure 4 for Global-Aware Enhanced Spatial-Temporal Graph Recurrent Networks: A New Framework For Traffic Flow Prediction
Viaarxiv icon

EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Masked Audio Gesture Modeling

Add code
Jan 02, 2024
Figure 1 for EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Masked Audio Gesture Modeling
Figure 2 for EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Masked Audio Gesture Modeling
Figure 3 for EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Masked Audio Gesture Modeling
Figure 4 for EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Masked Audio Gesture Modeling
Viaarxiv icon

Multi-Scale Spatial-Temporal Recurrent Networks for Traffic Flow Prediction

Add code
Oct 12, 2023
Viaarxiv icon

Exploring the Mutual Influence between Self-Supervised Single-Frame and Multi-Frame Depth Estimation

Add code
Apr 25, 2023
Viaarxiv icon

Attention-based Spatial-Temporal Graph Convolutional Recurrent Networks for Traffic Forecasting

Add code
Feb 25, 2023
Viaarxiv icon