- Dream, Lift, Animate: From Single Images to Animatable Gaussian Avatars (3DV 2026)
- GENMO: A Generative Model for Human Motion (ICCV 2025)
- GeoMan: Temporally Consistent Human Geometry Estimation using Image-to-Video Diffusion (ICCV 2025)
- AdaHuman: Animatable Detailed 3D Human Generation with Compositional Multiview Diffusion (ICCV 2025)
- HumanOLAT: A Large-Scale Dataset for Full-Body Human Relighting and Novel-View Synthesis (ICCV 2025)
- SimAvatar: Simulation-Ready Avatars with Layered Hair and Clothing (CVPR 2025)
- VideoPanda: Panoramic Video Diffusion with Multi-view Attention (arXiv 2025)
- COIN: Control-Inpainting Diffusion Prior for Human and Camera Motion Estimation (ECCV 2024)
- Multi-Track Timeline Control for Text-Driven 3D Human Motion Generation (ECCV 2024)
Research Experience
Currently a Senior Research Manager at NVIDIA Research, leading the Data-Drive AI for Robotics (DAIR) team. The team focuses on enabling robots to learn skills directly from human data.
Education
2014-2018: PhD in Computer Science from the University of Bonn, Germany, under the supervision of Prof. Juergen Gall. 2011-2013: Master's degree in Finland. 2006-2010: Bachelor's degree in Pakistan.
Background
Research Interests: Computer vision, machine learning, and robotics. Background: Senior Research Manager at NVIDIA Research, leading the Data-Drive AI for Robotics (DAIR) team. The team investigates how robots can learn directly from human data (e.g., videos, motion capture, and large-scale demonstrations) and generalize across tasks, embodiments, and environments.
Miscellany
Personal Interests: Passionate about computer vision and machine learning.