Institute of Robotics and Intelligent Systems D-MAVT

Open Opportunities

ANYmal - Navigation using self-supervised depth image representation ETH Zurich Robotic Systems Lab Project Objective: This project aims to develop a new method to integrate depth perception into robots. Specifically we are tackling the task of navigation and locomotion for our legged robot ANYmal. Instead of training RL policies from scratch, we would like to understand the potential performance impacts by using self-supervised training, e.g. using DINO, for monocular depth images. The training may consist of leveraging existing datasets and focuses on learning a useful representation that includes not only geometric hints but potentially also semantic information from depth analog to the recent advantages achieved using RGB images. Computer Hardware, Computer Perception, Memory and Attention, Computer Software, Computer Vision, Electrical and Electronic Engineering Master Thesis, Semester Project

Master Thesis / Internship / Semester Project: Digitization of large 12-lead ECG Image database ETH Zurich Spinal Cord Injury & Artificial Intelligence Lab Other organizations: Sensory-Motor Systems Lab 12-lead electrocardiograms (ECGs) are still solely documented on paper in many hospitals, especially in the Global South. These physical paper records provide a multitude of conditions recorded in many different countries. Our lab has access to a dataset with more than 8000 patient’s ECG photos / scans of 12-lead signals printed onto physical paper sheets. The dataset comprises 12-lead ECG image records from more than 35 hospital sites across Europe. The primary objective of this project is to develop an automated digitization pipeline from raw image scan in .png format towards 12 vectorized ECG time series in WFDB format. Computer Vision, Engineering and Technology, Medical and Health Sciences Bachelor Thesis, Internship, Master Thesis, Semester Project

Electrical Flow-Based Graph Embeddings for Event-based Vision and other downstream tasks University of Zurich Robotics and Perception This project explores a novel approach to graph embeddings using electrical flow computations. Artificial Intelligence and Signal and Image Processing, Knowledge Representation and Machine Learning, Mathematics Master Thesis

Autonomous shape morphing of robotic limbs for multi-task manipulation ETH Zurich Robotic Systems Lab We are working toward general-purpose quadruped robots that transform their functional shapes to adapt to new tasks and environments. This project will entail developing control policies in simulation and deploying them on hardware, with the goal of autonomously controlling a quadruped with shape-morphing limbs (see attached image a). Intelligent Robotics, Robotics and Mechatronics, Simulation and Modelling Master Thesis, Semester Project

What can Large Language Models offer to Event-based Vision? University of Zurich Robotics and Perception This project focuses on combining Large Language Models within the area of Event-based Computer Vision. Computer Vision Master Thesis, Semester Project

Hybrid Spiking-Deep Neural Network System for Efficient Event-Based Vision Processing University of Zurich Robotics and Perception Master's thesis in collaboration with SynSense: Neuromorphic Intelligence & Application Solutions Artificial Intelligence and Signal and Image Processing, Dynamical Systems Master Thesis

Learned Event Generation from Images University of Zurich Robotics and Perception In this project, the student applies concepts from current advances in image generation to create artificial events from standard frames. Multiple state-of-the-art deep learning methods will be explored in the scope of this project. Artificial Intelligence and Signal and Image Processing Master Thesis, Semester Project

Enhancing Robotic Motor Policies with Event Cameras University of Zurich Robotics and Perception The goal of this project is to develop a shared embedding space for events and frames, enabling the training of a motor policy on simulated frames and deployment on real-world event data. Artificial Intelligence and Signal and Image Processing Master Thesis, Semester Project

Event Keypoint Extraction for Real-Time Pose Estimation University of Zurich Robotics and Perception This project focuses on enhancing camera pose estimation by exploring a data-driven approach to keypoint extraction, leveraging recent advancements in frame-based keypoint extraction techniques. Artificial Intelligence and Signal and Image Processing Master Thesis, Semester Project

Autonomous Drone Navigation via Learning from YouTube Videos University of Zurich Robotics and Perception Inspired by how humans learn, this project aims to explore the possibility of learning flight patterns, obstacle avoidance, and navigation strategies by simply watching drone flight videos available on YouTube. Computer Vision Master Thesis