Projects are grouped into work I am actively pursuing and work that represents earlier phases.

Current Projects

Multi-Camera Multi-Sensor SLAM

A SLAM system that fuses multiple cameras and complementary sensors for robust localization and mapping across challenging motion and environment conditions.

360° SLAM

A panoramic SLAM system for mapping and localization using 360-degree imagery with robust loop closure and global consistency.

Image Based Gravity Estimation for VIO/SLAM

A current research project on calibrating noisy IMU gravity priors from a single RGB image for downstream VIO, SLAM, and visual localization systems.

Past Projects

Large-scale Indoor 3D Reconstruction from 360° Images

Panoramic indoor reconstruction for virtual tours using multi-view geometry, depth estimation, and scalable fusion.

Audio-driven Facial Animation Synthesis

Speech-driven facial animation with temporal modeling for realistic lip-sync and expression control.

ARSketch

A sketch-based user interface for augmented reality glasses that enables in-air sketching as a natural interaction mechanism.

Visual SLAM for AR Glasses

Real-time visual-inertial SLAM for resource-constrained AR devices with emphasis on robustness and efficiency.

HandAugment

A depth-based 3D hand pose estimation project combining two-stage hand region extraction and synthetic data augmentation based on MANO.

Deep Video Deinterlacing

A TensorFlow implementation of real-time deep video deinterlacing for reconstructing progressive frames from interlaced video.

Globally Optimal Toon Tracking

A SIGGRAPH 2016 paper on region correspondence and tracking for hand-drawn animation using global optimization over appearance and motion cues.