Papers for

uav operators

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Efficient aerial object detector improves small object recognition accuracy

YOLO12-MambaScan: An Efficient Object Detector with High-Frequency Enhancement and State-Space Modeling

Abstract: The rapid development of unmanned aerial vehicle (UAV) technology has made aerial-image object detection increasingly important for natural-resource monitoring, traffic management, and disaster response. Detecting small objects in aerial images remains difficult because objects occupy very few pixels, high-frequency cues are easily lost, and global context is hard to model in cluttered scenes. Existing detectors often retain insufficient edge, corner, and texture information. We propose \ours, an aerial-image detector built on the YOLO12 architecture. The model combines a triple-path high-frequency enhancement convolution module (TriPathHFConv), receptive-field coordinate-attention convolution (RFCAConv), and a Mamba-based global-context module. On VisDrone, at an input resolution of 960*960, ours achieves 60.0% mAP@50 and 38.6%mAP@50:95, demonstrating a favorable accuracy--efficiency trade-off for small-object detection. The benchmark and dataset protocol follow the VisDrone challenge setup.

Sat 12 SeptComputer Vision and Pattern RecognitionInformation Retrieval

The gist

Detecting tiny objects in drone images is hard because they take up very few pixels and details like edges and textures can get lost. The authors propose a new detector called YOLO12-MambaScan that enhances these high-frequency details and better captures the overall scene context. Their method improves the accuracy of finding small objects in aerial images while staying efficient enough to run at high resolutions. They tested it on a popular drone image dataset and showed better performance than existing detectors.

Open → 2609.13647v1