Weakly Supervised Training of Universal Visual Concepts for Multi-domain Semantic Segmentation

Deep supervised models have an unprecedented capacity to absorb large quantities of training data. Hence, training on multiple datasets becomes a method of choice towards strong generalization in usual scenes and graceful performance degradation in edge cases. Unfortunately, popular datasets often h...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	International journal of computer vision 2024-07, Vol.132 (7), p.2450-2472
Hauptverfasser:	Bevandić, Petra, Oršić, Marin, Šarić, Josip, Grubišić, Ivan, Šegvić, Siniša
Format:	Artikel
Sprache:	eng
Schlagworte:	Artificial Intelligence Computer Imaging Computer Science Datasets Image Processing and Computer Vision Labels Manholes Pattern Recognition Pattern Recognition and Graphics Performance degradation Roads Semantic segmentation Special Issue on Robust Vision Vans Vision
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!