Student Projects


How to apply

To apply, please send your CV, your Ms and Bs transcripts by email to all the contacts indicated below the project description. Do not apply on SiROP . Since Prof. Davide Scaramuzza is affiliated with ETH, there is no organizational overhead for ETH students. Custom projects are occasionally available. If you would like to do a project with us but could not find an advertized project that suits you, please contact Prof. Davide Scaramuzza directly to ask for a tailored project (sdavide at ifi.uzh.ch).


Upon successful completion of a project in our lab, students may also have the opportunity to get an internship at one of our numerous industrial and academic partners worldwide (e.g., NASA/JPL, University of Pennsylvania, UCLA, MIT, Stanford, ...).



Learning minimal representations of places - Available

Description: Place recognition and 6DoF localization has a wide range of applications, whether in robot autonomy, VR/AR or navigation interfaces. Given sensor readings (we focus on images), the goal is to establish position and orientation of a robot/device with respect to a previously recorded map. Recently, this is generally solved with a mixture of machine learning and geometry (NetVLAD, SuperPoint, LF-NET, PoseNet). Our focus in particular will be to solve this problem with a minimal representation.

Goal: Given query agent A and map agent B, have B establish a pose of A within its map, with minimal data transmission from A to B. We have a couple of ideas on how to solve this (see our most recent publication on this: https://arxiv.org/abs/1811.10681 ), but you are encouraged to bring your own ideas to the table.

Contact Details: Titus Cieslewski ( titus at ifi.uzh.ch ), ATTACH CV AND TRANSCRIPT (also Bachelor)! Preferred skills: Linux, Python, “Vision Algorithms for Mobile Robots” class or equivalent, TensorFlow/PyTorch or equivalent.

Thesis Type: Master Thesis

See project on SiROP

Autonomous Car Dataset with Event Cameras - Available

Description: Billions of dollars are spent each year to bring autonomous vehicles closer to reality. One of the remaining challenges is the design of reliable algorithms that work in a diverse set of environments and scenarios. At the core of this problem is the choice of sensor setup. Ideally, there is a certain redundancy in the setup while each sensor should also excel at a certain task. Standard cameras are an essential building block of any autonomous vehicle but they are not a good fit for high-dynamic range scenarios that frequently occur while driving.

Goal: In this project, we want to explore the utility of event cameras in an autonomous car scenario. In order to achieve this, a driving dataset will be created that incorporates not only common sensors such as standard cameras, GPS, IMU and possibly Lidar but also state-of-the-art event cameras. This project involves a diverse set of tasks including: - Creation of sensor setup - Design of a post-processing pipeline for convenient labeling - Data collection and selection

Contact Details: Mathias Gehrig (mgehrig at ifi.uzh.ch); Daniel Gehrig (dgehrig at ifi.uzh.ch)

Thesis Type: Master Thesis

See project on SiROP

Learn to Deblur Images with Events - Available

Description: Images suffer from motion blur due to long exposure in poor light condition or rapid motion. Unlike conventional cameras, event-cameras do not suffer from motion blur. This is due to the fact that event-cameras provide events together with the exact time when they were triggered. In this project, we will make use of hybrid sensors which provide both conventional images and events such that we can leverage the advantages of both.

Goal: The goal is to develop an algorithm capable producing a blur-free image from the captured, blurry image, and events within the exposure time. To this end, synthetic data can be generated by our simulation framework which is able to generate both synthetic event data and motion blurred images. This data can be used to train a machine learning algorithm designed to solve the task at hand. At the end of the project, the algorithm will be adapted to perform optimally with real-world data.

Contact Details: Mathias Gehrig, mgehrig (at) ifi (dot) uzh (dot) ch

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Safe Reinforcement Learning for Robotics - Available

Description: Reinforcement Learning (RL) has recently emerged has a technique to let robots learn by their own experience. Current methods for RL are very data-intensive, and require a robot to fail many times before actually accomplishing their goal. However some systems, such as flying robots, require to respect safety constraints during learning and/or deployment. While maximizing performance, those methods usually aim to minimize the number of system failures and overall risk.

Goal: During this project, we will develop machine learning based techniques to let a (real) drone learn to fly nimbly through gaps and gates, while minimizing the risk of critical failures and collisions.

Contact Details: **Antonio Loquercio** loquercio@ifi.uzh.ch

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Unsupervised Obstacle Detection Learning - Available

Description: Supervised learning is the gold standard algorithm to solve computer vision tasks like classification, detection or segmentation. However, for several interesting tasks (e.g. moving object detection, depth estimation, etc.) collecting the large annotated datasets required by the aforementioned algorithms is a very tedious and costly process. In this project, we aim to build a self-supervised depth estimation and segmentation algorithm by embedding classic computer vision principles (e.g. brightness constancy) into a neural network. **Requirements**: Computer vision knowledge; programming experience with python. Machine learning knowledge is a plus but it is not required.

Goal: The goal of this project consists of building a perception system which can learn to detect obstacles without any ground truth annotations.

Contact Details: Antonio Loquercio, _loquercio@ifi.uzh.ch_

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Optimization for Spiking Neural Networks - Available

Description: Spiking neural networks (SNNs) are neural networks that process information with timing of events/spikes rather than numerical values. Together with event-cameras, SNNs show promise to both lower latency and computational burden compared to artificial neural networks. In recent years, researchers have proposed several methods to estimate gradients of SNN parameters in a supervised learning context. In practice, many of these approaches rely on assumptions that might not hold in all scenarios. Requirements: - Background in machine learning; especially deep learning - Solid foundation in applied mathematics - Good programming skills; experience in CUDA is a plus.

Goal: In this project we explore state-of-the-art optimization methods for SNNs and their suitability to solve the temporal credit-assignment problem. As a first step, an in-depth evaluation of a selection of algorithms is required. Based on the acquired insights, the prospective student can propose improvements and implement their own method.

Contact Details: Mathias Gehrig, mgehrig (at) ifi (dot) uzh (dot) ch

Thesis Type: Master Thesis

See project on SiROP

High-Performance Simulation of Spiking Neural Network on GPUs - Available

Description: One major complication in research of biologically-inspired spiking neural Networks (SNNs) is simulation performance on conventional hardware (CPU/GPU). Computation in SNNs is dominated by operations on sparse tensors but usually this potential benefit is ignored to save development time. However, the exploitation of sparsity could be beneficial to scale simulation of SNNs to larger datasets. Requirements: - Experience with deep learning frameworks (e.g. TensorFlow or PyTorch) - Excellent programming skills and experience in CUDA

Goal: In this project, you will leverage sparse computation to develop high-performance simulations of SNNs that can be used for optimization. This will help to scale experiments and drastically improve results obtained by SNNs.

Contact Details: Mathias Gehrig, mgehrig (at) ifi (dot) uzh (dot) ch

Thesis Type: Semester Project

See project on SiROP

Target following on nano-scale UAV - Available

Description: Autonomous Unmanned Aerial Vehicles (UAVs) have numerous applications due to their agility and flexibility. However, navigation algorithms are computationally demanding, and it is challenging to run them on-board of nano-scale UAVs (i.e., few centimeters of diameter). This project focuses on the object tracking, (i.e., target following) on such nano-UAVs. To do this, we will first train a Convolutional Neural Network (CNN) with data collected in simulation, and then run the aforementioned network on a parallel ultra-low-power (PULP) processor, enabling flight with on-board sensing and computing only. **Requirements**: Knowledge of python, cpp and embedded programming. Machine learning knowledge is a plus but it is not strictly required.

Contact Details: Antonio Loquercio, _loquercio@ifi.uzh.ch_

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Decentralized Visual Map Building - Available

Description: In State-of-the-Art decentralized mapping methods, optimization (correcting odometry drift) is typically done using pose graph optimization due to the fact that a pose graph is a very compact representation. Unfortunately, this compression in data results in limitations in precision and robustness. Bundle Adjustment is a map optimization method for visual maps which is much more precise and robust, but also much more data intensive.

Goal: In this work, you will figure out a way to achieve the superior precision of Bundle Adjustment while minimizing the amount of data that needs to be exchanged between robots in a decentralized setting.

Contact Details: Titus Cieslewski ( titus at ifi.uzh.ch ), ATTACH CV AND TRANSCRIPT! Required skills: Matlab or C++, with a preference for the latter. Desirable: Background in optimization (Nonlinear least squares, Gauss-Newton or similar)

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Robust 3D exploration - Available

Description: In exploration, a robot creates a map of a previously unknown environment. Typically, the goal is to cover all of the reachable space. Think of a robot deployed in a disaster area, tasked with finding all survivors (search and rescue). The problem with most current exploration algorithms is that they assume perfect pose estimates. The problem is that robots equipped with on-board pose estimators will always produce an estimate with an error/drift. In this project, you will work on achieving exploration not with a perfect, but rather realistic pose estimate.

Goal: Explore an unknown space in 3D, relying only on visual-inertial odometry (with drift) and basic place recognition (but no loop closure/map optimization). Start in simulation, then possibly deploy in the real world (quad equipped with depth sensor).

Contact Details: Titus Cieslewski ( titus at ifi.uzh.ch ), ATTACH CV AND TRANSCRIPT (also Bachelor)! Required skills: Linux, ROS. Ideally C++ and Python, but if you know only one you should also be fine.

Thesis Type: Master Thesis

See project on SiROP

Event-based Deep Learning - Available

Description: Event cameras such as the Dynamic Vision Sensor (DVS) are recent sensors with large potential for high-speed and high-dynamic-range robotic applications. The goal of this project is to explore new algorithms for processing events within a deep-learning context. The goal of the project should be implementing and comparing different frameworks for processing events, and applying them to challenging tasks, such as optical flow prediction. This is a project with considerable room for creativity. Experience in coding image processing algorithms in C++ and experience with learning frameworks in python is required.

Goal: The goal of this project is to explore new algorithms for processing events within a deep-learning context.

Contact Details: Daniel Gehrig (dgehrig at ifi.uzh.ch), Antonio Loquercio (antonilo (at) ifi.uzh.ch)

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Asynchronous Processing for Event-based Deep Learning - Available

Description: Event cameras such as the Dynamic Vision Sensor (DVS) are recent sensors with large potential for high-speed and high dynamic range robotic applications. Since their output is sparse traditional algorithms, which are designed for dense inputs such as frames, are not well suited. The goal of this project is explore ways to adapt existing deep learning algorithms to handle sparse asynchronous data from events. Experience in C++ and python deep learning frameworks is required.

Goal: The goal of this project is explore ways to adapt existing deep learning algorithms to handle sparse asynchronous data from events.

Contact Details: Daniel Gehrig (dgehrig at ifi.uzh.ch)

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Open-Source Research Flight Controller interfacing Hobbyist Hardware - Available

Description: Due to the popularity of drones, a wide variety of models is available on the market. The branches of this market reaches from toys, consumer grade hardware for photography, professional equipment for cinematography to first-person-view drones for racing and freestyle flight, and even research drones. Although there are multiple research platforms available, they often come with disadvantages like high price, large size, high weight and a certain amount of ecosystem lock-in or restrictions on development tools. This project aims at programming existing flight controller hardware to replace the hobbyist software. Tasks within this project involve identifying the communication interfaces used, writing libraries for communication with motor controllers, on-board computers and sensors, and implementing simple filter and control functions. All of this runs embedded on an ARM chip (hardware will be provided) and in a POSIX environment. Good embedded programming skills are required.

Goal: The goal is a simple embedded flight stack, consisting of multiple libraries which should be easy to modify and maintain, and optionally could be released to the research community as a open-source package.

Contact Details: Philipp Föhn (foehn at ifi.uzh.ch)

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Generation of Fast or Time-Optimal Tracjectories for Quadrotor Flight - Available

Description: With the rise of complex control and planning methods, quadrotors are capable of executing astonishing maneuvers. While generating trajectories between two known poses or states is relatively simple, planning through multiple waypoints is rather complicated. The master class of this problem is the task of flying as fast as possible through multiple gates, as done in drone racing. While humans can perform such fast racing maneuvers at extreme speeds of more than 100 km/h, algorithms struggle with even planning such trajectories. Within this thesis, we want to research methods to generate such fast trajectories and work towards a time-optimal planner. This requires some prior knowledge in at least some of the topics including: planning for robots, optimization techniques, model predictive control, RRT, and quadrotors or UAVs in general. The tasks will reach from problem analysis, approximation, and solution concepts to implementation and testing in simulation with existing software tools.

Goal: The goal would be to analyse the planning problem, develop approximation techniques and solve it as time-optimal as possible during thesis.

Contact Details: Philipp Föhn (foehn at ifi.uzh.ch)

Thesis Type: Master Thesis

See project on SiROP

3D Reconstruction using an Event Camera - Available

Description: Event cameras such as the Dynamic Vision Sensor (DVS) are recent sensors with large potential for high-speed and high dynamic range robotic applications. In this project we want to explore the possibility of active sensing using an event camera, where the task is to do asynchronous 3D reconstruction. This is important for low-latency updates about the scene and can be useful in fast obstacle avoidance. Applicant should have a strong background in C++ programming and low-level vision.

Goal: In this project we want to explore the possibility of active sensing using an event camera, where the task is to do asynchronous 3D reconstruction.

Contact Details: Daniel Gehrig (dgehrig (at) ifi.uzh.ch), Dario Brescianini (brescianini (at) ifi.uzh.ch)

Thesis Type: Semester Project / Master Thesis

See project on SiROP

MPC for high speed trajectory tracking - Available

Description: Many algorithms exist for model predictive control for trajectory tracking for quadrotors and equally many implementation advantages and disadvantages can be listed. This thesis should find the main influence factors on high speed/high precision trajectory tracking such as: modell accuracy, aerodynamic forces modelling, execution speed, underlying low-level controllers, sampling times and sampling strategies, noise sensitivity or even come up with a novel implementation.

Goal: The end-goal of the thesis should be a comparison of the influence factors and based on that a recommendation or even implementation of an improved solution.

Contact Details: Philipp Föhn (foehn at ifi.uzh.ch)

Thesis Type: Master Thesis

See project on SiROP

Optical Flow Estimation with an Event Camera - Available

Description: Event cameras such as the Dynamic Vision Sensor (DVS) are recent sensors with large potential for high-speed and high dynamic range robotic applications. The goal of this project is to use event cameras to compute the optical flow in the image plane induced by either a moving camera in a scene or by moving objects with respect to a static event camera. Several existing methods as well as proposed new ones will be analyzed, implemented and compared. A successful candidate is expected to be familiar with state-of-the-art optical flow methods for standard cameras. This is a project with considerable room for creativity, for example in applying the ideas from low-level vision or ideas driving optical flow methods for standard cameras to the new paradigm of event-based vision. Experience in coding image processing algorithms in C++ is required.

Contact Details: Guillermo Gallego (guillermo.gallego at ifi.uzh.ch), Henri Rebecq (rebecq at ifi.uzh.ch)

Thesis Type: Master Thesis

See project on SiROP

Event Camera Characterization - Available

Description: Event cameras such as the Dynamic and Active Pixel Vision Sensor (DAVIS, http://inilabs.com/products/dynamic-and-active-pixel-vision-sensor/ ) are recent sensors with large potential for high-speed and high dynamic range robotic applications. In spite the successful demonstration of the sensor to address several problems in computer vision and robotics, a comprehensive characterization of the sensor for such high level applications is still missing.

Goal: The goal of this project is to characterize various aspects of these novel types of sensors, such as: event noise characteristics (distribution and spectral density), contrast threshold (relation to bias settings, variability: spatially, with the pixel, and photometrically, with respect to the scene illumination), non-linearities, etc. Additionally, the images and IMU measurements provided by the DAVIS also require an integrated characterization. A successful completion of the project will lead to a better understanding of the potential, limitations and impact of these sensors on the design of novel algorithms for computer vision and robotics. The expected candidate should have a background on instrumentation, electrical engineering (to understand the principle of operation of the DAVIS pixels) and random processes. This project involves close collaboration with the Institute of Neuroinformatics (INI) at UZH-ETH.

Contact Details: Guillermo Gallego (guillermo.gallego at ifi.uzh.ch)

Thesis Type: Semester Project / Bachelor Thesis

See project on SiROP

Building a high-speed camera! Learning Image reconstruction with an Event Camera - Available

Description: Event cameras such as the Dynamic Vision Sensor (DVS) are recent sensors with large potential for high-speed and high dynamic range robotic applications. The output of an event camera is a sparse stream of events that encode only light intensity changes - in other terms, a highly compressed version of the visual signal.

Goal: The goal of this project is to turn an event camera into a high-speed camera, by designing an algorithm to recover images from the compressed event stream. Inspired by a recent approach, the goal of this project will be to train a machine learning algorithm (or neural network) to learn how to reconstruct an image from the noisy event stream. The first part of the project will consist in acquiring training data, using both simulation and real event cameras. The second part will consist in designing and training a suitable machine learning algorithm to solve the problem. Finally, the algorithm will be compared against state-of-the-art image reconstruction algorithms. The expected candidate should have some background on both machine learning and computer vision (or image processing) in order to undertake this project.

Contact Details: Henri Rebecq (rebecq at ifi.uzh.ch), Guillermo Gallego (guillermo.gallego at ifi.uzh.ch)

Thesis Type: Master Thesis

See project on SiROP

A Visual-Inertial Odometry System for Event-based Vision Sensor - Available

Description: Event-based cameras are recent revolutionary sensors with large potential for high-speed and low-powered robotic applications. The goal of this project is to develop visual-inertial pipeline for the Dynamic and Active Vision Sensor (DAVIS). The system will estimate the pose of the DAVIS using the event stream and IMU measurements delivered by the sensor. Filtering approaches as well as batch optimization methods will be investigated. https://youtu.be/bYqD2qZJlxE http://www.inilabs.com/products/davis

Contact Details: Henri Rebecq (rebecq at ifi.uzh.ch), Guillermo Gallego (guillermo.gallego at ifi.uzh.ch)

Thesis Type: Master Thesis

See project on SiROP

Hand-Eye Calibration Toolbox - Available

Description: Hand-Eye calibration is a paramount pre-processing stage of many robotic and augmented reality applications, where the knowledge of the relative transformation between different sensors (e.g. a camera and a head-mounted display) is required to have an accurate geometric representation of the scene.

Goal: The goal of this project is to develop a user-friendly hand-eye calibration toolbox integrated with our robotic system. The toolbox will contain existing and novel hand-eye calibration methods, and it will allow to visualize the results of the different methods in an integrated manner to improve the understanding of the quality of the processed dataset, specially paying attention to error estimates, uncertainties and detection of inconsistent data.

Contact Details: Guillermo Gallego (guillermo.gallego at ifi.uzh.ch)

Thesis Type: Semester Project / Master Thesis

See project on SiROP

Integrated Multi-Camera Calibration Toolbox - Available

Description: The toolbox is expected to handle different camera brands, projection models and calibration patterns. In the multi-sensor scenario, the toolbox is also expected to compute the temporal offsets between the sensors. Special attention will be given to estimation of error measures, parameter uncertainties, detection of inconsistent data and interactive guidance of data acquisition.

Goal: The goal of this project is to develop a user-friendly, single and multi-camera calibration toolbox adapted to our robotic system. The toolbox will integrate existing calibration software in our group and in other libraries and will provide user-friendly reports of the different stages to assess the quality of the processed dataset, thus speeding up and improving the understanding of the whole sensor calibration stage.

Contact Details: Guillermo Gallego (guillermo.gallego at ifi.uzh.ch)

Thesis Type: Semester Project / Master Thesis

See project on SiROP