Publications

(2024). Motion Consistency Model: Accelerating Video Diffusion with Disentangled Motion-Appearance Distillation. NeurIPS'2024.

PDF Cite Code Poster ⭐️Project⭐️ ⭐️Demo⭐️ Checkpoint Colab

(2024). IDOL: Unified Dual-Modal Latent Diffusion for Human-Centric Joint Video-Depth Generation. ECCV'2024.

PDF Cite Code Poster Video ⭐️Project⭐️

(2024). Disco: Disentangled Control for Referring Human Dance Generation in Real World. CVPR'2024.

PDF Cite Code Slides ⭐️Project⭐️

(2023). Language-guided Human Motion Synthesis with Atomic Actions. ACM MM'2023.

PDF Cite Code Poster

(2023). SOAR: Scene-debiasing Open-set Action Recognition. ICCV'2023.

PDF Cite Code Poster Slides

(2023). High Fidelity 3D Hand Shape Reconstruction via Scalable Graph Frequency Decomposition. CVPR'2023.

PDF Cite Code Slides

(2022). Adaptive Two-Stream Consensus Network for Weakly-Supervised Temporal Action Localization. TPAMI'2022.

PDF Cite DOI

(2021). Action Coherence Network for Weakly-Supervised Temporal Action Localization. IEEE Transaction on Multimedia (TMM).

PDF Cite DOI ICIP Version

(2020). Two-Stream Consensus Network for Weakly-Supervised Temporal Action Localization. ECCV'2020.

PDF Cite DOI Supplement