CoRL 2026
Realistic Gaze Transformer (RGT), a Transformer-VAE framework that learns expressive full-head gaze dynamics from human motion.
I'm a first year Robotics PhD student at MBZUAI. Prior to that, I was a graduate research assistant in the Creative Machines Lab at Columbia University under Professor Hod Lipson.
I'm interested in the intersection of computer vision, deep learning, agentic AI, and robotics.
CoRL 2026
Realistic Gaze Transformer (RGT), a Transformer-VAE framework that learns expressive full-head gaze dynamics from human motion.
CVPR 2025
An unsupervised approach for understanding articulated motion and constructing description files for unseen articulated objects from 4D point cloud.
IEEE Robotics and Automation Letters (RA-L)
Neural Monocular SLAM enhanced with metric depth estimation and Gaussian-based unbounded scene representation.