University Tuebingen University of Ljubljana LOOKOUT Luxonis Shield AI Catskill AI Virginia Tech Dartmouth College Helmut-Schmidt-Universität Hamburg FBK KIT Southampton Oulu Lisboa CEVI KLE Tech BMWi

Maritime Computer Vision Workshop @ CVPR 2026

Keynote speakers and topics

Fiona Hua Parneet Kaur

From Sea to Street: What Maritime Autonomy Can Learn from Large-Scale AV Deployment

Fiona Hua (Einride) & Parneet Kaur

While the maritime industry is advancing rapidly, it faces the same scaling challenges that RoboTaxis and Autonomous freight transportation have spent years addressing. This presentation explores how maritime autonomy can accelerate development by adopting proven autonomous vehicle practices. Bridging the gap between sea and street, we unpack five critical lessons for scaling maritime autonomy, tracing the journey from defining product ODD, managing technical complexity at scale, impelemeting safety-driven verification and validation frameworks, addressing long-tail edge cases through tagged data curation, and deploying continuous fleet-learning infrastructure. Together, these capabilities form the foundation for scalable and commercially viable autonomous operations for maritime autonomy. By studying the AV industry's playbook, maritime stakeholders can bypass predictable pitfalls and chart a faster course to open water.
Marek Suchowski

AIS in the Wild: Coverage Gaps, Spoofing, and What CV Models Must Know

Marek Suchowski (catskill GmbH)

Automatic Identification System (AIS) data has become a cornerstone of maritime situational awareness. But its real-world reliability is far messier than most computer vision pipelines assume. We explore the hidden fragility of AIS as a ground-truth signal: where coverage breaks down, how spoofing and signal manipulation distort vessel tracks and what these failure modes mean for CV models trained or evaluated against such data.