Bachelor, Master and Doctoral School Projects

Doctoral School Projects

Master Thesis or Doctoral School Projects

For further project offers, please contact members of the CVLAB directly.

Semester Projects (Bachelor and Master)

Most of the offered semester projects can be rephrased for a thesis and vice versa. Please contact us directly.

Administrative

Semester Projects (Bachelor and Master)

SIN and SSC students do one semester project during their Bachelor studies and one semester project during their Master studies.

Semester projects can be done in groups of two students.

Semester projects are worth 8 credits for Bachelor and 12 credits for Master.

Students must have the approval of the Professor in charge of the laboratory before registering for the given project.

Oral defense: within two weeks of the hand-in date.

Master Thesis Projects

Master Thesis Projects are started once the complete master program is finished and all the credits have been obtained.

Projects for SSC and SIN students should last 4 months at the EPFL or 6 months in the industry or in another University.

Master Thesis Projects must be done individually.

Master Thesis Projects are worth 30 credits.

Students must have the approval of the Professor in charge of the laboratory before registering for the given project.

Additional information

This project aims to advance the field of object tracking by employing sophisticated decision algorithms like Monte Carlo Tree Search (MCTS) and its reinforcement learning variant, AlphaZero, to enhance the association of people detection across different frames or views.

Image from [4]. Comparison of different implicit network architectures fitting a ground truth image (top left). Second and third row show first and second order derivatives.

Implicit Neural Representations (INRs) have become increasingly popular across multiple tasks and domains, such as shape representation and novel view synthesis. Even more recently they have started to be adopted in image compression pipelines, opening a different direction compared to existing neural-based compression approaches, which need a training dataset and the consequent storage of a (…)

This project is dedicated to understanding how different anonymization techniques, particularly face blurring, affect the performance of detection models. By systematically blurring faces in training datasets and evaluating the resulting model accuracies, we aim to identify anonymization strategies that preserve the utility of the data while ensuring privacy.Project GoalsThe primary objectives of this project are:To (…)

Develop a comprehensive framework for generating near-photorealistic images that accurately represent deformable objects and changing materials.

Translation between French and Patois Fribourgeois of picturesque fribourgeois situation.

The project presented here aims to address these limitations by fine-tuning a Large Language Model (LLM) to the specific cultural and linguistic characteristics of the Canton de Fribourg.

Matching Visual Archives with Newspaper Events

Investigate algorithms that can extract information from both visual and textual content and link them together. This project aims to use Vision-Language Models to make connections. It will do this by using the models’ ability to understand and connect pictures with related contexts.

a pile of carton

Through this project, you will be developing a solution for a real-world application that will be used in the future within WasteFlow service and will help optimize recycling.

computer-vision object detector detecting trash items like cardboards and bottles on a conveyor belt

Through this project, you will be developing a solution for real-world application that will be used in the future within WasteFlow service and will help optimize recycling. 

Examples of 3D Medical Image Analysis beyond Voxels

Advanced deep learning methods to transcend this voxel-centric approach, thereby promoting a more efficient and precise interpretation of 3D medical images.

Five instances of Liver extracted from MRI data. The black dot indicates a pseudo key point.

Five instances of Liver extracted from MRI data. The black dot indicates a pseudo key point.In this project, we plan to explore methods to match/align shapes extracted from biomedical images, based on shape and texture-based features. By doing so, we will be able to analyze the variations of these structures over time and compare them (…)

Synthetic sTreak Rendering for sAtellite Kinematics and Surveillance

The goal of this project is to develop a tool that allows the insertion of realistic synthetic observations of space objects into astronomical images.

Spacecraft Dataset Development for Unseen and Occluded Targets

In this project, the student will construct a number of 3D spacecraft and debris objects, integrate a number tumble profiles, and insert these motions into various earth orbits. The student will then render images of the objects in orbit to create a public dataset.

DescriptionDifferent visual tasks are often strongly and obviously correlated. For instance, having surface normals simplifies estimating the depth of an image, knowing segmentation could help detect objects, etc. Our intuition implies the existence of a special structure among visual tasks. Extracting this structure would allow us to seamlessly reuse supervision among related tasks or solve (…)

Deep learning based approaches can be used for a wide range of space applications such as on-board data processing for observation satellite and collision prevention, spacecraft rendezvous, etc. Unfortunately, the deep learning models are very computational intensive and require huge amount of resources and power consumption. In recent years, some techniques like quantization, pruning and (…)

Computer Aided Engineering (CAE) is at the core of modern industrial engineering and manufacturing. However, the current CAE applications suffer from significant time and human resource expenses. Our goal is to leverage deep learning techniques to automate the CAE process and reduce the R&D costs for the industry.

Image from [4]. Comparison of different implicit network architectures fitting a ground truth image (top left). Second and third row show first and second order derivatives.

Implicit Neural Representations (INRs) have become increasingly popular across multiple tasks and domains, such as shape representation and novel view synthesis. Even more recently they have started to be adopted in image compression pipelines, opening a different direction compared to existing neural-based compression approaches, which need a training dataset and the consequent storage of a (…)

toroidal fusion reactor

An active research field is the prediction of physics mechanisms degrading plasma confinement and performance, eventually leading to disruptions on Tokamaks. Disruptions are the ultimate consequence of highly coupled non-linear plasma physics processes, resulting in an abrupt loss of plasma current and confinement inducing huge electromagnetic forces and thermal loads on Plasma Facing Components and (…)

Promoting Connectivity of Linear Structures in 3D Microscopy Images

As in many areas of computer vision, deep networks now deliver state-of-the-art results for delineation tasks, such as finding axons and dendrites in 3D light microscopy images. Most of the existing approaches rely on convolutional networks to extract from images binary masks denoting which voxels belong to neurites and which do not. Unfortunately, they do (…)