History Dependent Significance Coding for Incremental Neural Network Compression.
Gerhard TechPaul HaaseDaniel BeckingHeiner KirchhofferKarsten MüllerJonathan PfaffHeiko SchwarzWojciech SamekDetlev MarpeThomas WiegandPublished in: ICIP (2022)
Keyphrases
- neural network
- predictive coding
- image and video coding
- image and video compression
- arithmetic coding
- transform coding
- coding scheme
- shape coding
- progressive transmission
- efficient compression
- compression efficiency
- coding method
- compression algorithm
- block coding
- expert group
- compression scheme
- back propagation
- image compression
- artificial neural networks
- lossless coding
- coding gain
- bit allocation
- compression ratio
- bitstream
- image codec
- lossless compression
- data compression
- neural network model
- huffman coding
- run length encoding
- image coding
- bit rate
- embedded image
- genetic algorithm
- entropy coding
- neural network is trained
- incremental learning
- wavelet compression
- gray code
- video signals
- image coding algorithm
- bp neural network
- image quality
- bits per pixel
- fuzzy logic
- intra prediction
- vector quantizer
- inter frame
- variable bit rate
- embedded zerotree wavelet