Size & Shape Matters: The Need of HPC Benchmarks of High Resolution Image Training for Deep Learning.
Ferran Parés PontPedro MegiasDario Garcia-GasullaMarta Garcia-GasullaEduard AyguadéJesús LabartaPublished in: Supercomput. Front. Innov. (2021)
Keyphrases
- high resolution
- deep learning
- low resolution
- multiscale
- input image
- single image
- image features
- image representation
- super resolution
- image segmentation
- test images
- image retrieval
- deep architectures
- unsupervised learning
- image content
- image processing
- image classification
- segmentation algorithm
- training set
- segmentation method
- region of interest
- lighting conditions
- restricted boltzmann machine
- face images
- higher order
- supervised learning
- training data
- machine learning