MAST: Multiscale Audio Spectrogram Transformers.
Sreyan GhoshAshish SethS. UmeshDinesh ManochaPublished in: CoRR (2022)
Keyphrases
- multiscale
- image processing
- audio visual
- pattern analysis
- scale space
- wavelet transform
- multimedia
- multiscale analysis
- audio video
- natural images
- shape representation
- audio signal
- image representation
- filter bank
- audio signals
- image fusion
- visual data
- wigner distribution
- coarse to fine
- image segmentation
- multi modal
- signal processing
- edge detection
- audio stream
- wavelet domain
- music score
- digital audio
- multiscale representation
- multiresolution
- visual information
- affine invariant
- audio files
- image compression
- soccer video
- wavelet decomposition