Illustrates the structure of the C2F-SimDLKA module. After processing with CBS, the features are first split into two parts: one part is retained without any processing, and the other part is processed through several SimDLKA modules.

<p>Each SimDLKA module splits into two channels: one channel passes the processed features to the next SimDLKA module, while the other channel retains the features for later concatenation. Finally, after passing through n SimDLKA modules, all features are fused together.</p>

Saved in:

Bibliographic Details
Main Author:	Xunqian Xu (21261899) (author)
Other Authors:	Tao Wu (16763) (author), Zhongbao Du (21261902) (author), Hui Rong (5575562) (author), Siwen Wang (444607) (author), Shue Li (2873627) (author), Dakai Chen (21261905) (author)
Published:	2025
Subjects:	Neuroscience Science Policy Biological Sciences not elsewhere classified Information Systems not elsewhere classified novel attention mechanism human body behavior average frame rate approximately 3 frames 3 &# 8211 new loss function model &# 8217 pose estimation dataset original bottleneck module feature fusion module experimental results demonstrate traditional detection algorithms posture </ p human motion analysis yolov8 loss function original yolov8 model dciou loss function pose detection loss functions traditional yolov8 model training results indicate yolov8 framework simdlka module valuable reference thereby improving source coco significant interest selectively focus mpii datasets method provides map value input data independently constructed forward propagation detecting poses crucial task c2f module also suffer
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!

Illustrates the structure of the C2F-SimDLKA module. After processing with CBS, the features are first split into two parts: one part is retained without any processing, and the other part is processed through several SimDLKA modules.

Similar Items