Lightweight SAR Ship Detection via Contrastive Distillation
Quick Take
The proposed SURGE framework enables efficient SAR ship detection by transferring relational geometry from a teacher detector to a compact student detector using contrastive InfoNCE objectives, achieving up to 6.2 mAP and 8.0 AP75 improvements on SSDD and HRSID benchmarks. This marks the first transformer-based knowledge distillation approach in the SAR domain, enhancing performance without altering existing model architectures.
Key Points
- SURGE framework uses contrastive InfoNCE for knowledge distillation in SAR detection.
- Achieves significant performance gains of 6.2 mAP and 8.0 AP75 on SSDD and HRSID.
- First transformer-based knowledge distillation framework for SAR ship detection.
- Architecture-agnostic, compatible with various detector models without modifications.
- Improves efficiency for real-time and onboard SAR ship detection applications.
Article Content
From source RSS / original summaryarXiv:2605. 30380v1 Announce Type: new Abstract: 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
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →Evi-Steer: Learning to Steer Biomedical Vision-Language Models through Efficient and Generalizable Evidential Tuning
Evi-Steer introduces a novel evidential tuning framework for BiomedCLIP, enabling efficient fine-tuning with only 0.11% parameter updates. It significantly enhances performance in few-shot learning and domain shifts across 15 biomedical imaging datasets, demonstrating robustness for clinical applications.
