DINO-Detect: A Simple yet Effective Framework for Blur-Robust AI-Generated Image Detection

2025-11-16

AI Summary

Motion blur—arising from handheld capture or video compression—severely degrades the performance of existing AI-generated image (AIGI) detectors in real-world scenarios. To address this, we propose a teacher-student knowledge distillation framework specifically designed for motion-blur-robust AIGI detection. We freeze a high-capacity self-supervised teacher model (DINOv3) and leverage its rich semantic features and logit responses extracted from sharp images to supervise a lightweight student model trained directly on blurred images. Crucially, our method introduces dual-granularity distillation—jointly operating at both feature-level and logit-level—without requiring additional blur modeling or image enhancement. Evaluated on diverse synthetic multi-scale motion blur and realistic degraded datasets, our approach consistently outperforms state-of-the-art methods, achieving up to an 8.2% absolute improvement in detection accuracy. This demonstrates superior generalization across blur types and strong practical viability for real-world deployment.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

With growing concerns over image authenticity and digital safety, the field of AI-generated image (AIGI) detection has progressed rapidly. Yet, most AIGI detectors still struggle under real-world degradations, particularly motion blur, which frequently occurs in handheld photography, fast motion, and compressed video. Such blur distorts fine textures and suppresses high-frequency artifacts, causing severe performance drops in real-world settings. We address this limitation with a blur-robust AIGI detection framework based on teacher-student knowledge distillation. A high-capacity teacher (DINOv3), trained on clean (i.e., sharp) images, provides stable and semantically rich representations that serve as a reference for learning. By freezing the teacher to maintain its generalization ability, we distill its feature and logit responses from sharp images to a student trained on blurred counterparts, enabling the student to produce consistent representations under motion degradation. Extensive experiments benchmarks show that our method achieves state-of-the-art performance under both motion-blurred and clean conditions, demonstrating improved generalization and real-world applicability. Source codes will be released at: https://github.com/JiaLiangShen/Dino-Detect-for-blur-robust-AIGC-Detection.

Problem

• Detecting AI-generated images under motion blur degradation
• Improving detector robustness in real-world handheld photography
• Addressing performance drops from texture distortion in blur

Innovation

• Teacher-student knowledge distillation for blur-robust detection
• Frozen DINOv3 teacher provides stable semantic representations
• Distillation enables consistent performance under motion blur degradation