|
Drew Scheffer I am a PhD student in the AI Lab at the University of Michigan advised by Prof. Stella Yu. I also earned my BSE (2020-2023) and MSE (2023-2024) at the University of Michigan. Previously, I interned at the NASA Jet Propulsion Laboratory where I worked on the surface simulation software for the Mars 2020 mission, Mars Sample Return mission, and the soon-to-be-launched CADRE tech demo. Please feel free to reach out with any questions, inquiries, or jokes via my (scrambled) email below :) hvi{womw@yqmgl.ihy -- |
|
ResearchI'm broadly interested in robotics and computer vision. At the moment, I'm particularly interested in interactive navigation and active perception. |
|
VIGOR: Visual Goal-In-Context Inference for Unified Humanoid Fall Safety
Osher Azulay, Zhengjie Xu, Drew Scheffer*, Stella X. Yu In Submission project page / arxiv A teacher-student framework that distills terrain-aware recovery goals into a deployable egocentric-depth policy. |
|
ShipwreckFinder: A QGIS Tool for Shipwreck Detection in Multibeam Sonar Data
Anja Sheppard*, Tyler Smithline*, Drew Scheffer, David Smith, Advaith V. Sethuraman, Ryan Bird, Sabrina Lin, Katherine A. Skinner OCEANS, 2025 arxiv / github We introduce ShipwreckFinder, an open-source QGIS plugin that detects shipwrecks from multibeam sonar data. |
|
MorphEUS: Morphable OmnidirEctional Unmanned System
Ivan Bao*, José C. Díaz Peón González Pacheco*, Atharva Navsalkar*, Drew Scheffer*, Sashreek Shankar*, Andrew Zhao*, Hongyu Zhou*, Vasileios Tzoumas ICRA Workshop, 2025 (*Ordered Alphabetically) project page / pdf / arxiv MorphEUS is a compact, morphable co-axial quadrotor that enables efficient, resilient omnidirectional flight with independent position and orientation control, making it ideal for close-range inspection and manipulation tasks. |
|
Wreck-tify: Consistent Diffusion Correction For Shipwreck Mapping
Drew Scheffer, Ashwin Saxena, Christian Foreman Class Project, 2024 presentation / github / pdf Our team presents various diffusion-based underwater image restoration methods that are used to dehaze and color-correct underwater images such that they are conducive to above-water SLAM and scene reconstruction pipelines. |
|
Da-TRASH: Depth-appended Tabletop Recycling Algorithm
for Segmenting Havoc
Drew Scheffer, Ashwin Saxena Class Project, 2023 project page / github / pdf Our group extends upon previous work by adapting an existing Unseen Object Instance Segmentation (UOIS) model so that it can perform segmentation on RGB-only images. |
|
Treble in the Sheets: Optical Music Recognition
Drew Scheffer, Ashwin Saxena, Christian Foreman Class Project, 2022 video / github / pdf This work presents a simplified optical musical recognition system. Our system takes images of sheet music as input and outputs both an image of the annotated sheet music and a MIDI file for easy parsing |
Game Development |
|
Sub-merged Conflicts
Christian Foreman, Drew Scheffer, Adam Scovera, Manish Venumuddula Collaborative 7-week game design project where you and a friend are underwater explorers on a mission to find the treasure, navigate through treacherous currents, encounter dangerous creatures, and use your teamwork skills to keep your submarine safe and strike gold! |
|
Periodic Puffins
Drew Scheffer A video game prototype by where you play as a pixelated puffin who’s hungry for FISH. However, like in life, you’re not in complete control… |
|
Zelda -- The Legend of Duck
Christian Foreman, Drew Scheffer Video game design project that implements the Legend of Zelda's first dungeon in addition to a custom level! |
Teaching |
|
Based on this template. |