3D Reconstruction Research · ASU Logos Lab

01 / OVERVIEW

Context & problem

As a Research Aide at the ASU Logos Lab, developed Python/PyTorch transformer models for 3D reconstruction and articulated physical models. The research aims to help robots interact physically with digital twins reconstructed from images.

A useful digital twin needs more than a plausible static shape: it must represent articulated structure in a way that can support downstream physical interaction.

02 / SYSTEM ARCHITECTURE

How the pieces connect.

03 / INDIVIDUAL CONTRIBUTION

What I worked on.

• Developed transformer-based research models in Python and PyTorch.
• Worked on 3D reconstruction and articulated physical-model representations.
• Connected image-based reconstruction research to robotics interaction goals.

04 / IMPLEMENTATION

Engineering details.

A useful digital twin needs more than a plausible static shape: it must represent articulated structure in a way that can support downstream physical interaction.

• Research focused on transformer architectures and image-conditioned 3D representations.
• Experiments were conducted as university research, not a commercially deployed system.

05 / RESULTS & VALIDATION

What the work supports.

• Developed research models and experimentation workflows; no publication or benchmark is claimed.

Validation details are limited to evidence stated in the resume; no unreported datasets, accuracy claims, or benchmark comparisons are added.