Object detection from aerial platforms under adverse atmospheric conditions, particularly haze, is paramount for robust drone autonomy. Yet, this domain remains largely underexplored, primarily hindered by the absence of specialized benchmarks. To bridge this gap, we present \textit{HazyDet}, the first, large-scale benchmark specifically designed for drone-view object detection in hazy conditions. Comprising 383,000 real-world instances derived from both naturally hazy captures and synthetically hazed scenes augmented from clear images, HazyDet provides a challenging and realistic testbed for advancing detection algorithms. To address the severe visual degradation induced by haze, we propose the Depth-Conditioned Detector (DeCoDet), a novel architecture that integrates a Depth-Conditioned Kernel to dynamically modulate feature representations based on depth cues. The practical efficacy and robustness of DeCoDet are further enhanced by its training with a Progressive Domain Fine-Tuning (PDFT) strategy to navigate synthetic-to-real domain shifts, and a Scale-Invariant Refurbishment Loss (SIRLoss) to ensure resilient learning from potentially noisy depth annotations. Comprehensive empirical validation on HazyDet substantiates the superiority of our unified DeCoDet framework, which achieves state-of-the-art performance, surpassing the closest competitor by a notable +1.5\% mAP on challenging real-world hazy test scenarios. Our dataset and toolkit are available atthis https URL.

View on arXiv

@article{feng2025_2409.19833,
  title={ HazyDet: Open-Source Benchmark for Drone-View Object Detection with Depth-Cues in Hazy Scenes },
  author={ Changfeng Feng and Zhenyuan Chen and Xiang Li and Chunping Wang and Jian Yang and Ming-Ming Cheng and Yimian Dai and Qiang Fu },
  journal={arXiv preprint arXiv:2409.19833},
  year={ 2025 }
}

Comments on this paper