← Toutes les offres
DU

Postdoctoral Research Associate, Department of Computer Science

DURHAM UNIVERSITY

Durham
CDD
Education
Publié le 19 novembre 2026

Description du poste

The Role and Department The Department of Computer Science at Durham University invites applications to join an ambitious interdisciplinary collaboration at the interface of computer vision, generative artificial intelligence and biodiversity conservation. The successful candidate will work in an internationally connected research environment, with direct links to the Royal Botanic Gardens, Kew and botanical experts in Madagascar. The role is primarily based in Durham; flexible or hybrid arrangements may be considered subject to research, computing and collaborative requirements and University policies. The Role Applications are invited for a Postdoctoral Research Associate in Computer Vision, Image Processing and Generative AI to contribute to CHLORIS (Computational Herbarium Learning for Optical Reconstruction and Identification Systems). The project investigates how digitised herbarium specimens can be transformed into scientifically useful illustrations of living plants, retaining the morphological characters needed for reliable species identification. It brings together Durham University's machine-learning expertise and the Royal Botanic Gardens, Kew's botanical and conservation expertise. Research at Durham will be advised by Professor Paolo Remagnino, working closely with Dr Stuart Cable and collaborating botanists, Kew. We seek an outstanding researcher and exceptionally capable programmer who can translate ideas into robust, efficient and reproducible research software. The appointee will develop image analysis and pattern recognition pipelines to segment specimens, extract leaf morphology and venation, and quantitatively assess visual fidelity; they will also adapt modern generative image models using structural, semantic and botanical conditioning. The post combines substantive methodological research with practical delivery of tested code, datasets, benchmarks and publishable findings. Prior botanical expertise is welcome but not required.