Lightweight SAR Ship Detection via Contrastive Distillation
This work provides a method to create more efficient and lightweight SAR ship detection models, which is crucial for real-time or onboard deployment in SAR applications.
This paper addresses the computational cost of SAR ship detection models by proposing a knowledge distillation framework called SURGE. It transfers relational geometry from a large teacher model to a compact student model using a contrastive InfoNCE objective, achieving up to 6.2 mAP and 8.0 AP75 gains over baseline student models on SSDD and HRSID benchmarks, and even surpassing teacher performance.
Deep convolutional and transformer-based detectors achieve strong performance for SAR ship detection but are often computationally prohibitive for real-time or onboard deployment. Lightweight models offer improved efficiency yet struggle to capture the complex structural relationships inherent in SAR backscatter. Most existing SAR knowledge-distillation approaches rely on feature or logit matching, which enforces localized activation similarity while neglecting the geometric relationships among object representations. We propose a Structured Unified Relational knowledGE distillation framework for SAR Ship detection (SURGE) that transfers relational geometry from a powerful teacher detector to a compact student detector using a contrastive InfoNCE objective in a shared projection embedding space. To the best of our knowledge, this work presents the first transformer-based SAR ship detector knowledge distillation framework in SAR domain. The framework is architecture-agnostic in the sense that it provides a common region-level distillation interface for two-stage, one-stage and transformer-based detectors without modifying their deployed architectures. Experiments on the SSDD and HRSID benchmarks demonstrate that the proposed method yields substantial improvements for two-stage detectors, achieving up to 6.2 mAP and 8.0 AP75 gains over baseline student and even surpassing teacher performance