7 papers
Quality-Preserving Imperceptible Adversarial Attack on Skeleton-based Human Action Recognition
Ziyi Chang, Kanglei Zhou, Xiaohui Liang +1
Adversarial attacks on skeletal human action recognition have received significant attention. However, existing methods typically introduce noise-like perturbations that degrade mo…
Physics-Based Motion Tracking of Contact-Rich Interacting Characters
Xiaotang Zhang, Ziyi Chang, Qianhui Men +1
Motion tracking has been an important technique for imitating human-like movement from large-scale datasets in physics-based motion synthesis. However, existing approaches focus on…
ART: Adaptive Relational Transformer for Pedestrian Trajectory Prediction with Temporal-Aware Relations
Ruochen Li, Ziyi Chang, Junyan Hu +3
Accurate prediction of real-world pedestrian trajectories is crucial for a wide range of robot-related applications. Recent approaches typically adopt graph-based or transformer-ba…
Motion In-Betweening for Densely Interacting Characters
Xiaotang Zhang, Ziyi Chang, Qianhui Men +1
Motion in-betweening is the problem to synthesize movement between keyposes. Traditional research focused primarily on single characters. Extending them to densely interacting char…
Real-time and Controllable Reactive Motion Synthesis via Intention Guidance
Xiaotang Zhang, Ziyi Chang, Qianhui Men +1
We propose a real-time method for reactive motion synthesis based on the known trajectory of input character, predicting instant reactions using only historical, user-controlled mo…
On the Design Fundamentals of Diffusion Models: A Survey
Ziyi Chang, George Alex Koulieris, Hyung Jin Chang +1
Diffusion models are learning pattern-learning systems to model and sample from data distributions with three functional components namely the forward process, the reverse process,…