My Resume

About Me

I am a PhD candidate at ETH Zürich’s Sensing, Interaction & Perception Lab (SIPLAB), working on machine learning for egocentric human motion estimation and understanding. My research focuses on body-worn sensing, including sparse inertial sensors, ultra-wideband ranging, and egocentric multi-view cameras, with the goal of capturing and interpreting human motion in the wild. Before my PhD I completed my Master’s at ETH with a major in Machine Intelligence and my Master’s Thesis at MIT CSAIL on 3D semantic segmentation from 2D supervision.

Publications

Ultra Diffusion Poser: Diffusion-Based Human Motion Tracking from Sparse Inertial Sensors and Ranging-based Between-sensor Distances

Dominik Hollidt, Tommaso Bendinelli, Christian Holz

CVPR 2026

Group Inertial Poser: Multi-Person Pose and Global Translation from Sparse Inertial Sensors and Ultra-Wideband Ranging

Yifeng Xue, Jiaxi Jiang, Rayan Armani, Dominik Hollidt, Yi-Chi Liao, Christian Holz

ICCV 2025

EgoSim: An Egocentric Multi-View Simulator and Real Dataset for Body-Worn Cameras during Motion and Activity

Dominik Hollidt, Paul Streli, Jiaxi Jiang, Yasaman Haghighi, Chuhan Qian, Xucong Liu, Christian Holz

NeurIPS 2024

Geometry Aware Field-to-field Transformations for 3D Semantic Segmentation

Dominik Hollidt, Clinton Wang, Polina Golland, Marc Pollefeys

2023

Education

PhD Candidate

December 2023 - now

SIPLAB, ETH Zürich - Zürich, CH

Researching machine learning for egocentric human motion estimation and understanding. Focus areas include body-worn sensing with sparse inertial sensors, ultra-wideband ranging, and egocentric multi-view cameras for capturing and interpreting human motion in the wild.

Human Motion Estimation
Egocentric Perception
Sensor Fusion

Master Thesis

March 2023 - August 2023

MIT (CSAIL, Medical Vision Group - Prof. Polina Golland) - Cambridge, USA

Worked on deep 3D semantic segmentation from 2D supervision only. Utilized neural radiance fields and field-to-field transformation via point cloud transformers, and investigated the effects of pretraining on fine-tuning performance for 3D semantic segmentation.

Master Studies

September 2020 - November 2023

ETH Zürich - Zürich, CH

MSc in Computer Science with a major in Machine Intelligence and a minor in Theoretical Computer Science.

Machine Intelligence
Theoretical Computer Science

Exchange Student

August 2019 - June 2020

KTH Royal Institute of Technology - Stockholm, SE

Completed 90 ECTS with straight A’s. Wrote my Bachelor Thesis on Optical Flow Algorithms for Event-Based Cameras (graded 1.3 German score), including a real-world optical flow dataset for event-based cameras to study the impact of resolution across different optical flow algorithms.

Machine Learning
Computer Vision
Robotics

Bachelor Studies

September 2017 - June 2020

Technical University Berlin - Berlin, GER

BSc in Computer Science. Graduated in the top 3% with a 1.3 German score.

Internships and Projects

Selected industry internships, research collaborations, and student projects.

Daedalean (06.2022 - 10.2022) - Visual traffic detection internship. Built synthetic flight scenarios from historical flight data to evaluate intruder detection accuracy, performed statistical analysis of historical flight tracks, reduced the sim-to-real gap via automatic color grading, and contributed to the data evaluation pipeline and an automated upload station for flight test data.

AMZ Driverless, ETH Formula Student (09.2020 - 09.2022) - Developed LiDAR-based cone detection (CNN) and multi-LiDAR calibration for an autonomous racing car, built CI/CD infrastructure (Jenkins, Docker) and a custom KPI tool, and created a photorealistic Unreal Engine simulator for end-to-end autonomous driving validation. Served as Business Team Lead. 3rd place at FS Germany Driverless.

KTH Formula Student (08.2019 - 06.2020) - Perception team of the autonomous racing car project. Implemented fast point cloud clustering for LiDAR-based cone detection and contributed to team coordination.

AVM (06.2020 - 09.2020) - Data analysis internship. Linked customer support data to crashes and bugs, built automated data summaries with Spark and Zeppelin, analyzed WiFi speed tests via statistical tests, and built a decision-tree-based tool for surfacing bug-causing configurations.

Selected student projects - A handful of smaller things I've enjoyed building on the side over the years: a Wikipedia alt-text annotation tool for accessibility, weather transfer between images via GANs (Deep Learning course), drug-candidate prediction with transformers on knowledge graphs, an AR memory game for Android (Unity / Vuforia) accessible to visually impaired players, free-text semantic search in PDFs using pretrained language models, a music-reactive remote-controlled LED strip, and an automatic cat feeder.