Postdoc in Computer Vision and AI for Marine Image Analysis at DTU Compute
Job Description
DTU Compute’s Visual Computing section is hiring a postdoc for the HAV-INNO project, which turns underwater video into a quantitative tool for measuring species abundance in Danish seas. The main task is estimating ‘MaxN’ (the largest number of one species visible in any single frame) across a large video collection, using deep learning for detection, counting and related problems.
At a glance
- Post: Postdoc
- Section: Visual Computing, DTU Compute (HAV-INNO project)
- Duration: 2 years
- Apply by: 21 September 2026
- Apply: 21 September 2026 (23:59 Danish time)
What the post involves
- Develop deep learning computer vision methods for underwater video
- Estimate MaxN counts to measure species abundance
- Research object detection, counting and anomaly detection
- Work with marine biologists, and supervise student projects if you wish
Why this post matters
Counting fish by hand in hours of video is a major bottleneck for marine monitoring. AI that does it reliably would change how ecosystems are measured, and the post combines strong computer vision research with real environmental impact and access to DTU’s high-performance computing resources.
What DTU is looking for
- A Ph.D. in computer vision, machine learning or a related field
- Strong deep learning skills
- Interest in applying AI to ecology
Nice to have
- Experience with video analysis or counting tasks
Pay and location
Salary follows the collective agreement with the Danish Confederation of Professional Associations, with supplements negotiable under DTU’s salary structure for scientific staff; the ad does not give a figure.
How to apply
Apply through the official DTU page. The closing date is 21 September 2026 (23:59 Danish time). Apply online through DTU’s recruitment site with the documents the ad lists (usually a CV, cover letter, degree certificates and, for postdocs, a list of publications).
See all our postdoc jobs, or browse more research jobs on ResearchJobs.in.
Hiring institution: Technical University of Denmark
Official advertisement: efzu.fa.em2.oraclecloud.com
How to prepare for this application
- Detection and counting: show relevant models you built.
- Video: temporal or tracking work helps.
- Interdisciplinary: the project works with biologists.
- Deadline: 21 September 2026.
About DTU
The Technical University of Denmark (DTU), in Kongens Lyngby north of Copenhagen, is one of Europe's leading technical universities. Its departments include DTU Electro (photonics and electrical engineering), DTU Compute, DTU Construct, DTU Physics and DTU Wind and Energy Systems, and its DTU Nanolab cleanroom is shared across the university. English is the working language in research.