Multi-Resolution Dual-Tree Wavelet Scattering Network for Signal Classification

TL;DR

This paper presents a Deep Scattering network using Dual-Tree complex wavelets for signal classification, achieving improved accuracy by extracting translation-invariant features across multiple resolutions.

Contribution

It introduces a novel multi-resolution Dual-Tree wavelet scattering network that enhances translation invariance and outperforms existing methods on various datasets.

Findings

Outperforms Mallat's ScatterNet on four datasets

Extracts translation-invariant, densely spaced signal representations

Improves classification accuracy across different modalities

Abstract

This paper introduces a Deep Scattering network that utilizes Dual-Tree complex wavelets to extract translation invariant representations from an input signal. The computationally efficient Dual-Tree wavelets decompose the input signal into densely spaced representations over scales. Translation invariance is introduced in the representations by applying a non-linearity over a region followed by averaging. The discriminatory information in the densely spaced, locally smooth, signal representations aids the learning of the classifier. The proposed network is shown to outperform Mallat's ScatterNet on four datasets with different modalities on classification accuracy.