Patch and Model Size Characterization for On-Device Efficient-ViTs on Small Datasets Using 12 Quantitative Metrics

Vision transformers (ViTs) have emerged as a successful alternative to convolutional neural networks (CNNs) in deep learning (DL) applications for computer vision (CV), particularly excelling in accuracy on large-scale datasets within high-performance computing (HPC) or cloud domains. However, in th...

Full description

Bibliographic Details

Published in:	IEEE Access
Main Authors:	Jurn-Gyu Park, Aidar Amangeldi, Nail Fakhrutdinov, Meruyert Karzhaubayeva, Dimitrios Zorbas
Format:	Article
Language:	English
Published:	IEEE 2025-01-01
Subjects:	Deep learning efficient vision transformers (ViTs) edge-AI on-device ML multi-objective optimization characterization
Online Access:	https://ieeexplore.ieee.org/document/10858160/