PhD - Multi-Modal Foundatin Models for Earth Observation
Mission
As a PhD student, you will investigate the potential of large-scale deep learning models, namely foundation models, for mining information from multi-modal Earth observation (EO) data. This will require the design and development of scientific methodologies for: 1) EO-oriented foundation models considering the heterogeneous multi-sensor nature of remote sensing images; 2) the integration of the advances in generic multi-modal foundation models in EO; and 3) the adaptation of EO-focused foundation models for environmental understanding through end-use cases with transfer learning. These efforts will contribute to a European Commission-supported Horizon project, which involves many partners distributed across Europe. The project aims at advancing generic multi-modal foundation model development over various data modalities to solve many tasks in different domains, including EO.
Main duties and responsibilities
You will be expected to:
- Design deep learning-based methodologies for the EO-focused foundation models development, considering the highly heterogeneous characteristics of remote sensing sensors, spanning spectral range, radiometric resolution, spatial resolution, and temporal coverage.
- Integrate the advances in generic multi-modal foundation models to EO by bridging the gap across various data modalities and domains.
- Apply downstream transfer of large-scale pretraining to EO-focused end-use cases for environmental understanding (e.g., species distribution modelling, flood management, land-use and land-cover mapping, etc.).
- Write publications.
- Attend international conferences, both in computer vision and remote sensing.
- Participate in the educational activities of the ENAC faculty.
Profile
- You have a master’s degree in computer science, electrical or environmental engineering or equivalent.
- You have experience with remote sensing, machine learning, and computer vision.
- You are proficient in Python coding.
- You are fluent in English.
- You are motivated, curious, and willing to work in a highly dynamic team.
We offer
- An opportunity to develop a scientific career in an exciting area of science.
- A unique opportunity to learn high-end techniques and approaches.
- Excellent educational conditions and competitive remuneration.
- A multi-cultural and stimulating scientific environment, in the EPFL-Valais campus in Sion.
Informations
The selection process involves multiple stages, and short-listed candidates will be requested to apply to a specific EPFL Doctoral Program to qualify for a PhD at EPFL, i.e. [EDCE Civil and Environmental Engineering doctoral program].
Please check this page for additional information on admission.
Please note that this is a separate application process necessary to be eligible to complete your PhD at EPFL.
Please send the following application documents in one single pdf-file:
- A letter of interest (max. 2 pages)
- An updated CV
- Contact details of 3 reference persons.
- Scans of your diplomas and grades from all academic institutions of higher education (after and not including high-school) listed in your CV. If you do not yet have your Master’s diploma, please send a certificate issued by your current institution of higher education that confirms your enrolment in a Master’s program).
- Copy of your passport or official identity document.
- A sample of your published work (if available), Master’s thesis or semester project.
Your application will only be considered if these documents are appended in one single pdf-file in the above stated order.
Application deadline: May, 10th 2025
Contract Start Date: July, 1st 2025
Activity Rate Min: 100,00
Activity Rate Max: 100,00
Contract Type: CDD
Duration: 1 year (candidacy examination) + up to 5 years after
Reference: 1538