CV

Robot Software Engineer

Polybot

Training and deploying diffusion-based generative policies and real-time, distributed multi-object tracking algorithms for farming tasks, and benchmarks on the Spot mini robot.

• Generative AI
• Tracking

Researcher

Embodied Vision, Max Planck Institute for Intelligent Systems

Developed and published novel computer vision methods for 3D rigid body tracking from RGB-D video data, incorporating physically plausible modeling of rigid body dynamics with an accompanying video dataset with motion capture ground truth and physics annotations. Research on deep learning techniques for 6D pose estimation of rigid objects from RGB images, learning-based planning and control through contact/frictional constraints using learned physics models and differentiable rendering for unsupervised learning of scene physics from images.

• Generative AI
• Tracking

Masters

University of Siegen

Mechatronics Engineering

• Robotics
• Control
• Optimization
• Computer Vision
• Machine Learning

Bachelors

National Institute of Technology

Mechanical Engineering

• Kinematics
• Dynamics
• Mechanical Design
• Thermodynamics
• Mechatronics
• Optimization

Physics-Based Rigid Body Object Tracking and Friction Filtering From RGB-D Videos

3DV, 2024

Physical Representation Learning and Parameter Identification from Video Using Differentiable Physics

IJCV, 2021

Learning to Identify Physical Parameters from Video Using Differentiable Physics

DAGM-GCPR, 2020