A Novel Approach to Flexible Multi-Resolution Image Compression using Deep Learning Based Autoencoders on Overlapping Image Patch.
David Liemanto GunawanYosi KristianPublished in: ICCSCI (2023)
Keyphrases
- deep learning
- image patches
- multiresolution
- restricted boltzmann machine
- deep belief networks
- natural images
- denoising
- weakly supervised
- sparse representation
- unsupervised feature learning
- superpixels
- unsupervised learning
- machine learning
- test images
- low resolution
- sparse coding
- feature vectors
- orientation histogram
- object class
- mental models
- super resolution
- data sets
- image representation
- data analysis
- face recognition