YOLOH: You Only Look One Hourglass for Real-Time Object Detection

Multi-scale detection based on Feature Pyramid Networks (FPN) has been a popular approach in object detection to improve accuracy. However, using multi-layer features in the decoder of FPN methods entails performing many convolution operations on high-resolution feature maps, which consumes signific...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on image processing 2024-01, Vol.33, p.2104-2115
Hauptverfasser:	Wang, Shaobo, Chen, Renhai, Wu, Hongyue, Li, Xiaozhe, Feng, Zhiyong
Format:	Artikel
Sprache:	eng
Schlagworte:	Accuracy Computational modeling Convolution Feature extraction Feature maps MISO communication Monolayers Multilayers multiscale fusion Object detection Object recognition One-stage object detection Real time semantic extraction Semantics single-shot network Training
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!