Computer Vision and Geometry Group

Open Opportunities

Floating Hands ETH Zurich Computer Vision and Geometry Group The idea of the "Floating Hands" project is to allow one to create holographic tutorials easily. The application then provides holographic instructions for users of the HoloLens performing a specific task. This could be thought of as the HoloLens version of YouTube How-To videos. Information, Computing and Communication Sciences Semester Project

An Augmented Arthroscopy System using Machine Learning Methods for Surgical Guidance University of Zurich Research in Orthopedic Computer Science Other organizations: Computer Vision and Geometry Group, Snedeker Group / Laboratory for Orthopaedic Biomechanics, pd|z Product Development Group Zurich This work aims to enhance minimally invasive joint surgery using Machine Learning Computer Graphics, Computer Software, Computer Vision, Computer-Human Interaction, Engineering and Technology, Image Processing, Intelligent Robotics, Neural Networks, Genetic Alogrithms and Fuzzy Logic, Pattern Recognition, Virtual Reality and Related Simulation Master Thesis

Goal-driven Exploration of Unknown Spaces using Landmarks and Semantic Relationships ETH Zurich Computer Vision and Geometry Group The goal of this project is to develop a vision-based learning method that allows an autonomous agent to explore a previously unknown space with the goal of discovering a queried semantic. This task can become particularly complex in unknown large-scale environments. A way to decrease the complexity, path traveled, and time needed to complete the task, is to provide the agent with a prior understanding about how the world is structured. For example, when asked to discover all the chairs in an apartment, the agent can benefit from prior knowledge of where it is most probable to find the queried semantic (e.g., next to a desk or a dining table). Our premise is that, if the agent can identify while exploring the space a larger and more prominent landmark that is commonly found close to the queried semantic, it will complete the task faster and having traveled a shorter path. In this project, we will leverage object co-occurences and other semantic relationships during learning to perform goal-driven exploration in simulation using existing indoor real-world datasets. As part of the project, we will test generalization of the method in several unseen environments, hence it is important to devise a method that does not rely on the metric space. Computer Vision, Robotics and Mechatronics Master Thesis, Semester Project, Summer School

Semantic SLAM with Quality-adaptive Properties ETH Zurich Computer Vision and Geometry Group We do not need everything in anything. Reconstructing our 3D world is not an exception. In many applications (e.g., construction progress monitoring) we only need to reconstruct certain semantics in high fidelity (quality) — e.g., only walls or chairs; the rest of the world can be reconstructed in lower quality. The goal of this project is to devise a semantic Simultaneous Localization and Mapping (SLAM) method that reconstructs the world in a quality-adaptive manner; which means it only reconstructs with high fidelity a queried semantic, leaving the rest in a lower level of fidelity. Computer Vision, Intelligent Robotics Master Thesis, Semester Project, Summer School

Image-based Localization in a Known Environment and Map Update in a Quality-adaptive Manner ETH Zurich Computer Vision and Geometry Group The goal of the project is to localize a set of sensors (e.g., cameras) mounted to a moving vehicle (e.g., robot) in a given 3D map of the environment while continuously updating it. Being quality-adaptive means two things in this case. First, the method should progressively increase the localization and reconstruction accuracy until the uncertainty falls below a threshold. Second, the algorithm should cope with maps built in a quality-adaptive manner - the reconstruction accuracy and detailedness changes at different locations of the map. Computer Vision, Intelligent Robotics Master Thesis, Semester Project, Summer School

Recovering Blurry Human Body ETH Zurich Computer Vision and Geometry Group When humans move, the images are usually corrupted by motion blur. All previous methods for human body and shape reconstruction assume sharp images. In this thesis, the goal is to design a method that could deal with motion blur. Computer Graphics, Computer Vision, Image Processing, Signal Processing, Virtual Reality and Related Simulation Master Thesis