Convolutional autoencoder pretraining plus deep filter banks and Fisher vector pooling yields improved texture classification accuracy and lower complexity than transformer-based masked autoencoders on standard texture datasets.
Title resolution pending
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
-
A self-supervised learning approach to deep filter banks for texture recognition
Convolutional autoencoder pretraining plus deep filter banks and Fisher vector pooling yields improved texture classification accuracy and lower complexity than transformer-based masked autoencoders on standard texture datasets.