Keep Clear of the Edges : An Empirical Study of Artificial Intelligence Workload Performance and Resource Footprint on Edge Devices.
Kun SuoTu N. NguyenYong ShiJing Selena HeChih-Cheng HungPublished in: IPCCC (2022)
Keyphrases
- artificial intelligence
- edge information
- edge detector
- weighted graph
- edge map
- edge pixels
- edge detection
- step edges
- resource usage
- undirected graph
- edge detection algorithm
- edge enhancement
- multiple scales
- edge orientation
- image edges
- edge direction
- edge localization
- edge linking
- edge segments
- response time
- canny edge
- resource allocation
- edge weights
- edge points
- expert systems
- mobile devices
- resource management
- multiscale
- detected edges
- edge strength
- directed graph
- gradient field
- computer science
- edge directed
- resource constraints
- sharp edges
- high frequency components
- subpixel accuracy
- disk space
- gradient magnitude
- directional filters
- intelligent systems
- resource consumption
- homogeneous areas
- zero crossing
- canny edge detection
- machine learning
- junction detection
- gradient direction
- image gradient
- natural language processing
- software engineering
- adjacent pixels
- corner points
- image prior
- photometric properties
- directed acyclic graph
- ambient intelligence
- image processing