Computer vision algorithms and hardware implementations: A survey
Алгоритмы компьютерного зрения и их аппаратные реализации: обзор
2019-08-06
SCID: 54.1/e6ku64qg
Discuss with AI
computer visiondeep learninghardware acceleratorsimage classificationobject detection
Figures from the paper
Abstract (AI)
The field of computer vision is experiencing a great-leap-forward development today. This paper aims at providing a comprehensive survey of the recent progress on computer vision algorithms and their corresponding hardware implementations. In particular, the prominent achievements in computer vision tasks such as image classification, object detection and image segmentation brought by deep learning techniques are highlighted. On the other hand, review of techniques for implementing and optimizing deep-learning-based computer vision algorithms on GPU, FPGA and other new generations of hardware accelerators are presented to facilitate real-time and/or energy-efficient operations. Finally, several promising directions for future research are presented to motivate further development in the field.
Key Findings
1
Deep learning has produced prominent achievements in image classification, object detection, and image segmentation.
2
GPU, FPGA, and emerging hardware accelerators enable optimization of deep-learning-based vision algorithms for real-time and energy-efficient operation.
3
The paper identifies promising future research directions for continued development of computer vision algorithms and hardware.
4
The survey emphasizes the importance of co-optimizing algorithms and hardware to support practical computer vision deployment.
5
The survey reviews recent advances in computer vision algorithms and corresponding hardware implementations.
Research Object
deep-learning-based computer vision algorithms and their hardware implementations
Research Subject
recent progress, implementation and optimization for image classification, object detection, and image segmentation, including real-time and energy-efficient operation
Publication Details
Publication Date
2019-08-06
Journal
Publisher
ISSN
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest