• 3D simulation super-resolution achieving high-resolution facial animation 115x faster (at 18 FPS) than traditional method while maintaining similar quality.
• Enhances low-resolution simulations to high-resolution surfaces while generalizing to unseen facial expressions, external forces, and dynamics.

• Proposes DuoVAE, a dual-VAE framework that probabilistically links an auxiliary VAE to a controllable latent space, enabling robust property-controlled generation even under noisy labels.
• Generates diverse synthetic training data for medical imaging, mitigating limited annotated data for downstream deep learning.

• Introduces a novel motion capture system in a multi-camera setup using a specially designed low-cost suit, enabling the accurate 3D reconstruction of a moving human body.
• Uses deep neural networks and geometric algorithms for accurate corner detection, labeling, and 3D reconstruction.
• Developed and deployed the multi-camera calibration framework enhanced by machine learning-based anomaly corner detection.

• Introduces 3D hand rendering framework in VR for wearable fingertip tracking devices, focusing on reconstructing realistic hand poses using only fingertip positions.
• Key contributions include motion retargeting and new hinge constraints for real-time inverse kinematics to enhance visual plausibility and precision.

• Introduces a VR dental simulator that renders highly stiff haptic feedback, achieving up to 10× greater maximum stiffness (Z-width) from virtual teeth using commercially available haptic devices.

• Introduces a stretchable, bi-functional skin-like thermo-haptic (STH) device for VR, capable of precise cold and hot sensations with a single structure and 230% stretchability.

• Proposes laser-assisted dual-function copper nanowire (CuNW) polyurethane acrylate (PUA) patterns as feedback-controllable stretchable heaters for a 12-pixel thermal haptic device, with enhanced mechanical and chemical durability.
• Demonstrates a stretchable thermal haptic device on a nylon glove that replicates heat transfer in various virtual environments, achieving temperature control with less than 5% error.

• Introduces a remote multiuser collaboration system via wearable cutaneous haptic interfaces, leveraging passivity-based simulations for stable and realistic interactions.