Multi-Task Learning-Enabled Automatic Vessel Draft Reading for Intelligent Maritime Surveillance
In this work, we solved a practical engineering problem, automatic vessel draft reading (VDR), with the intelligent vision technology. Specifically, we designe a multi-task learning network to complete different subtasks in VDR, and we constrcuct a novel dataset for the network training.
Jingxiang Qu
PhD
My research interests include Equivariant Learning, multimodal/graph learning, and their application to solve real-world problems.