![[2602.13297] Conditional Generative Models for High-Resolution Range Profiles: Capturing Geometry-Driven Trends in a Large-Scale Maritime Dataset](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.13297] Conditional Generative Models for High-Resolution Range Profiles: Capturing Geometry-Driven Trends in a Large-Scale Maritime Dataset
Summary
This paper explores the use of conditional generative models to synthesize high-resolution range profiles (HRRPs) for maritime surveillance, emphasizing the importance of geometric factors in robust HRRP generation.
Why It Matters
The research addresses the challenges of radar automatic target recognition in maritime environments, where acquisition conditions can significantly affect performance. By leveraging a large-scale dataset, the findings provide insights into improving the robustness of HRRP generation, which is crucial for effective coastal surveillance.
Key Takeaways
- Conditional HRRP generation can enhance radar target recognition.
- Geometric factors, such as ship dimensions and aspect angles, are critical for effective HRRP synthesis.
- Utilizing a large-scale maritime dataset allows for more robust model training and validation.
- The study demonstrates the potential of generative models in addressing variability in operational scenarios.
- Understanding acquisition geometry is essential for improving HRRP generation.
Computer Science > Computer Vision and Pattern Recognition arXiv:2602.13297 (cs) [Submitted on 9 Feb 2026] Title:Conditional Generative Models for High-Resolution Range Profiles: Capturing Geometry-Driven Trends in a Large-Scale Maritime Dataset Authors:Edwyn Brient (CMM), Santiago Velasco-Forero (CMM), Rami Kassab View a PDF of the paper titled Conditional Generative Models for High-Resolution Range Profiles: Capturing Geometry-Driven Trends in a Large-Scale Maritime Dataset, by Edwyn Brient (CMM) and 2 other authors View PDF Abstract:High-resolution range profiles (HRRPs) enable fast onboard processing for radar automatic target recognition, but their strong sensitivity to acquisition conditions limits robustness across operational scenarios. Conditional HRRP generation can mitigate this issue, yet prior studies are constrained by small, highly specific datasets. We study HRRP synthesis on a largescale maritime database representative of coastal surveillance variability. Our analysis indicates that the fundamental scenario drivers are geometric: ship dimensions and the desired aspect angle. Conditioning on these variables, we train generative models and show that the synthesized signatures reproduce the expected line-of-sight geometric trend observed in real data. These results highlight the central role of acquisition geometry for robust HRRP generation. Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) Cite as: arXiv:2602.13297 [cs.CV]...
