FNA++: Fast Network Adaptation via Parameter Remapping and Architecture Search

Deep neural networks achieve remarkable performance in many computer vision tasks. Most state-of-the-art ( SOTA ) semantic segmentation and object detection approaches reuse neural network architectures designed for image classification as the backbone, commonly pre-trained on ImageNet. However, per...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on pattern analysis and machine intelligence 2021-09, Vol.43 (9), p.2990-3004
Hauptverfasser:	Fang, Jiemin, Sun, Yuzhu, Zhang, Qian, Peng, Kangjian, Li, Yuan, Liu, Wenyu, Wang, Xinggang
Format:	Artikel
Sprache:	eng
Schlagworte:	Ablation Adaptation Artificial neural networks Computer architecture Computer vision Computing costs Fast network adaptation Image classification Image segmentation neural architecture search Neural networks Object detection Object recognition parameter remapping Parameters Pose estimation Search problems Searching Semantic segmentation Semantics Task analysis
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!