TL;DR
WUTDet is a large-scale, diverse ship detection dataset with benchmarks that facilitate the evaluation and development of detection algorithms in complex maritime environments.
Contribution
The paper introduces WUTDet, a comprehensive dataset with 100K images and diverse scenarios, and provides systematic benchmarks of multiple detection architectures.
Findings
Transformer models outperform CNNs in detection accuracy and small-object detection.
CNN models are more efficient for real-time ship detection.
Models trained on WUTDet generalize well across different datasets.
Abstract
Ship detection for navigation is a fundamental perception task in intelligent waterway transportation systems. However, existing public ship detection datasets remain limited in terms of scale, the proportion of small-object instances, and scene diversity, which hinders the systematic evaluation and generalization study of detection algorithms in complex maritime environments. To this end, we construct WUTDet, a large-scale ship detection dataset. WUTDet contains 100,576 images and 381,378 annotated ship instances, covering diverse operational scenarios such as ports, anchorages, navigation, and berthing, as well as various imaging conditions including fog, glare, low-lightness, and rain, thereby exhibiting substantial diversity and challenge. Based on WUTDet, we systematically evaluate 20 baseline models from three mainstream detection architectures, namely CNN, Transformer, and Mamba.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
