Structure of SEAM network [16].

<div><p>In UAV aerial photography scenarios, targets exhibit characteristics such as multi-scale distribution, a high proportion of small targets, complex occlusions, and strong background interference. These characteristics impose high demands on detection algorithms in terms of fine-gr...

Full description

Saved in:

Bibliographic Details
Main Author:	Ju Liang (4277062) (author)
Other Authors:	Fan Wang (135182) (author), Jia Chen (8203) (author), Hai-Yan Huang (20524427) (author), Zu-Fan Dou (22693557) (author)
Published:	2025
Subjects:	Sociology Plant Biology Space Science Biological Sciences not elsewhere classified Chemical Sciences not elsewhere classified Information Systems not elsewhere classified universal inverted bottleneck strong background interference spatial feature recalibration single semantic representation mean average precision grained feature extraction experimental results show enhancing feature integration efficient feature selection dynamic weighted fusion 1 percentage points original sppf module key target regions feature extraction module existing advanced algorithms enhancement attention module target scale characteristics targets exhibit characteristics small target detection scale fusion capability occluded small targets small targets detection algorithms scale targets occluded regions module focuses limited capability c3k2_uib module c3k2 module scale distribution scale connections detection stability yolov11s model xlink "> visdrone2019 dataset unoccluded areas tinyperson dataset thereby improving specifically compensates significant limitations scales related practical applications occlusion resistance information loss increases parameters high proportion effectively addressing complex occlusions adaptive bifpn 34 %,
Tags:	Add Tag No Tags, Be the first to tag this record!

Structure of SEAM network [16].

Similar Items