Sequential encoder-decoder distillation provides robust initialization for lightweight image compression autoencoders, followed by standard training that preserves reconstruction quality better than direct optimization in early epochs.
To support deployment on hardware-constrained devices, we adopt a reduction strategy with lower computa- tional cost in training time and dataset size
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Efficient training for compact compression models via sequential distillation
Sequential encoder-decoder distillation provides robust initialization for lightweight image compression autoencoders, followed by standard training that preserves reconstruction quality better than direct optimization in early epochs.