Deep image compression based on multi-scale deformable convolution.
Daowen LiYingming LiHeming SunLu YuPublished in: J. Vis. Commun. Image Represent. (2022)
Keyphrases
- image compression
- multiscale
- wavelet transform
- image processing
- vector quantization
- subband
- compression scheme
- compression algorithm
- deformable models
- low bit rate
- reconstructed image
- edge detection
- image representation
- scale space
- wavelet decomposition
- natural images
- compression ratio
- image segmentation
- wavelet coefficients
- discrete cosine transform
- image coding
- image transmission
- coarse to fine
- local binary pattern
- compressed images
- multiple scales
- deep learning
- entropy coding
- compression standards
- transform coding
- computer vision
- embedded zerotree
- optic flow
- multiresolution
- lossless compression
- diffusion equation
- fast fourier transform
- keypoints
- super resolution
- fine scale
- linear filtering
- feature extraction
- convolution kernel
- block truncation coding
- feature selection