Generative Bayesian Inference for Remote Sensing Inverse Problems (GENESIS-RS)

About the Project

Project Description

Remote sensing measurements provide indirect, noisy, and often incomplete observations of environmental systems. Tasks such as super-resolution, SAR despeckling, and spectral unmixing are fundamentally ill-posed inverse problems, yet most modern deep learning approaches focus on producing point estimates and lack reliable uncertainty quantification. This limits their scientific interpretability and practical usefulness in environmental monitoring and decision-making.

This PhD project aims to develop generative Bayesian inference frameworks for remote sensing inverse problems, enabling principled posterior estimation and uncertainty-aware reconstruction of environmental variables from satellite data. The project will position modern generative AI models as learned Bayesian priors, integrated with physically motivated observation models to solve inverse problems in a probabilistic and interpretable manner.

Aims and Methods

The core aims of the project are to:

1. formulate key remote sensing inverse problems within a Bayesian framework;
2. develop expressive generative priors using modern AI models; and
3. perform posterior inference that yields both reconstructions and calibrated uncertainty estimates.

Methodologically, the project will combine:

• Bayesian inverse problem formulations with explicit likelihood models derived from sensor characteristics (e.g. multiplicative speckle noise for SAR, resolution degradation operators for optical imagery);
• Generative models such as conditional diffusion models, normalizing flows, or flow-matching approaches to represent high-dimensional image priors;
• Posterior inference techniques, including variational inference and posterior sampling, to approximate or sample from the Bayesian posterior;
• Rigorous uncertainty evaluation using posterior predictive checks and calibration diagnostics.

The project will focus on one or two representative applications, such as SAR despeckling or multispectral super-resolution using Sentinel-1 and Sentinel-2 data, ensuring feasibility within the PhD timeframe while maintaining methodological depth.

Deliverables

• Novel Bayesian formulations for selected remote sensing inverse problems
• Algorithms for uncertainty-aware image reconstruction using generative priors
• Open-source software and reproducible experimental pipelines
• Peer-reviewed publications in machine learning and remote sensing venues

Keywords

Bayesian inference; inverse problems; generative models; diffusion models; normalizing flows; uncertainty quantification; remote sensing; environmental imaging; Sentinel data

Contact for information on the project:

Dr Oktay Karakuş – karakuso@cardiff.ac.uk

How to Apply

This project is accepting applications all year round, for self-funded candidates.

Mode of Study: Full-time or part-time

Please submit your application via Computer Science and Informatics - Study - Cardiff University

In the funding field of your application, indicate “I am applying for a self-funded PhD in Computer Science and Informatics”, and specify the project title and supervisors of this project in the text box provided.

Academic criteria: A 2:1 Honours undergraduate degree or a master's degree, in computing or a related subject. Applicants with appropriate professional experience are also considered. Degree-level mathematics (or equivalent) is required for research in some project areas.

Applicants must demonstrate English language proficiency. Students who do not have English as a first language must prove this by obtaining an IELTS score of at least 6.5 overall, with a minimum of 6.0 in each skills component. A full list of accepted qualifications is available here: https://www.cardiff.ac.uk/study/international/english-language-requirements/postgraduate

If you are interested, please contact Dr Oktay Karakus (karakuso@cardiff.ac.uk) sending your CV in the first instance. The application process requires you to develop an individual research proposal jointly with the supervision team, which builds on the information provided in this advert.

Once you have developed the proposal with support from the supervisors, please submit your application following the instructions provided below.

Please submit your application via Computer Science and Informatics - Study - Cardiff University

In order to be considered candidates must submit the following information:

• In the ‘Research Proposal’ section of the application enter the name of the project you are applying to and upload your Individual research proposal. Your research proposal should not exceed 2000 words, including references and bibliography.
• A personal statement (as part of the university application form, or as a separate attachment, if you prefer).
• A CV. Guidance on CVs for a PhD position can be found on the FindAPhD website.
• Qualification certificates and Transcripts - original and English translation, if applicable.
• References x 2 which should be academic references. Please note you need to provide the reference documents as part of your application.
• Proof of English language (if applicable).

Interview– If the application meets all of the entrance requirements listed above, you will be invited to an interview.