학술논문

Tree-Based Unidirectional Neural Networks for Low-Power Computer Vision
Document Type
Periodical
Source
IEEE Design & Test IEEE Des. Test Design & Test, IEEE. 40(3):53-61 Jun, 2023
Subject
Computing and Processing
Components, Circuits, Devices and Systems
Measurement
Convolutional neural networks
Visualization
Computer vision
Task analysis
Memory management
Horses
Language
ISSN
2168-2356
2168-2364
Abstract
Editor’s notes: Convolutional neural networks (CNNs) have paved paths for highaccuracy computer vision. This article presents a tree-based hierarchy of multiple shallow CNNs to enable their low-power implementations for embedded devices. —Muhammad Shafique, New York University Abu Dhabi