Yuanhao Zhai

PhD student@University at Buffalo·美国·OpenReview

研究方向

Open-set action recognition · Image manipulation detection · Action recognition · Temporal action localization · Weakly-supervised temporal action localization

7.5

SlowFast-VGen: Slow-Fast Learning for Action-Driven Long Video Generation

ICLR 2025Spotlight

5.0

EndoAssistant: A Large-scale Vision-Language Dataset for Endoscopic Surgery Understanding from Open-Source Videos

ICLR 2025Rejected

三作

6.0

Motion Consistency Model: Accelerating Video Diffusion with Disentangled Motion-Appearance Distillation

NeurIPS 2024Poster

一作

4.0

Compensating for Nonlinear Reduction with Linear Computations in Private Inference

ICLR 2024Rejected

二作

4.0

DisCo: Disentangled Control for Realistic Human Dance Generation

合作者 (20)