Performance comparison between the proposed method and YOLO-based state-of-the-art methods on the SeaDronesSee dataset under varying model complexities.

<p>(a) mAP curves for different parameter scales; (b) mAP curves for different computational complexities.</p>

محفوظ في:

التفاصيل البيبلوغرافية
المؤلف الرئيسي:	Beigeng Zhao (21637197) (author)
مؤلفون آخرون:	Jiawen Zhao (6580982) (author), Rui Song (331963) (author), Lizhi Yu (12476136) (author), Xia Zhang (41398) (author), Jiren Liu (12901496) (author)
منشور في:	2025
الموضوعات:	Biochemistry Space Science Biological Sciences not elsewhere classified Information Systems not elsewhere classified rarely occurring objects rapidly detecting objects propose targeted optimizations modifications significantly improve extra detection head experimental evaluations conducted demonstrated significant progress aware upsampling algorithm proposed model achieves lightweight complexity range div >< p based maritime search enhance yolo11 ’ relative accuracy improvement model ’ maritime search yolo series specific task rescue tasks rescue operations recent advancements method renowned mambayolo across lightweight configurations lightweight architecture include integrating high accuracy heavyweight configurations enhanced yolo11 detect small depth module common challenges code supporting captured images art methods
الوسوم:	إضافة وسم لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!

مواد مشابهة