Reducing Training Data Using Pre-Trained Foundation Models: A Case Study on Traffic Sign Segmentation Using the Segment Anything Model

The utilization of robust, pre-trained foundation models enables simple adaptation to specific ongoing tasks. In particular, the recently developed Segment Anything Model (SAM) has demonstrated impressive results in the context of semantic segmentation. Recognizing that data collection is generally...

Descrición completa

Detalles Bibliográficos

Principais autores:	Sofia Henninger, Maximilian Kellner, Benedikt Rombach, Alexander Reiterer
Formato:	Artigo
Idioma:	Inglês
Publicado:	MDPI AG 2024-09-01
Series:	Journal of Imaging
Assuntos:	semantic segmentation segment anything model Mask R-CNN training data reduction traffic signs
Acceso en liña:	https://www.mdpi.com/2313-433X/10/9/220
Tags:	Engadir etiqueta Sen Etiquetas, Sexa o primeiro en etiquetar este rexistro!