Bridging the Scale Gap: Balanced Tiny and General Object Detection in Remote Sensing Imagery

2025-12-01

AI Summary

To address the cross-scale detection performance imbalance caused by the coexistence of densely packed small objects and large objects in remote sensing imagery, this paper proposes ScaleBridge-Det—the first large-scale detection framework specifically designed for tiny object detection. It innovatively adapts large model architectures to remote sensing tiny object detection by introducing two key components: (1) a Routing-Enhanced Mixture Attention (REM) module that enables scale-adaptive expert routing, and (2) a Density-Guided Dynamic Query (DGQ) module that generates density-aware dynamic queries. These modules jointly alleviate multi-scale feature competition and uneven computational resource allocation. Evaluated on AI-TOD-V2 and DTOD benchmarks, ScaleBridge-Det achieves state-of-the-art performance; it further demonstrates strong cross-domain robustness on VisDrone, significantly improving tiny-object detection accuracy and overall scale balance.

Technology Category

• Computer Vision: Object Detection & Categorization
• Search and Optimization: Learning to Search
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Tiny object detection in remote sensing imagery has attracted significant research interest in recent years. Despite recent progress, achieving balanced detection performance across diverse object scales remains a formidable challenge, particularly in scenarios where dense tiny objects and large objects coexist. Although large foundation models have revolutionized general vision tasks, their application to tiny object detection remains unexplored due to the extreme scale variation and density distribution inherent to remote sensing imagery. To bridge this scale gap, we propose ScaleBridge-Det, to the best of our knowledge, the first large detection framework designed for tiny objects, which could achieve balanced performance across diverse scales through scale-adaptive expert routing and density-guided query allocation. Specifically, we introduce a Routing-Enhanced Mixture Attention (REM) module that dynamically selects and fuses scale-specific expert features via adaptive routing to address the tendency of standard MoE models to favor dominant scales. REM generates complementary and discriminative multi-scale representations suitable for both tiny and large objects. Furthermore, we present a Density-Guided Dynamic Query (DGQ) module that predicts object density to adaptively adjust query positions and numbers, enabling efficient resource allocation for objects of varying scales. The proposed framework allows ScaleBridge-Det to simultaneously optimize performance for both dense tiny and general objects without trade-offs. Extensive experiments on benchmark and cross-domain datasets demonstrate that ScaleBridge-Det achieves state-of-the-art performance on AI-TOD-V2 and DTOD, while exhibiting superior cross-domain robustness on VisDrone.

Problem

• Balances detection across tiny and large objects in remote sensing imagery
• Addresses scale variation and density distribution challenges in detection
• Optimizes performance for dense tiny objects without sacrificing general object detection

Innovation

• Scale-adaptive expert routing for multi-scale features
• Density-guided dynamic query allocation for objects
• Balanced detection across tiny and large objects