Phase-Aware Wavelet-Based-Scattering Encoder-Decoder for Dense Predictions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Marrakchi, Ghassen, Matei, Basarab
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917528130813952
author Marrakchi, Ghassen
Matei, Basarab
author_facet Marrakchi, Ghassen
Matei, Basarab
contents Scattering transforms achieve Lipschitz stability and translation invariance, but dense prediction tasks require preserving spatial structure lost in global averaging. We propose Phase-Aware Scattering Encoder-Decoder, which restores this information by explicitly preserving phase in skip connections. On image denoising (BSD68), breaking translation invariance improves PSNR by $+2.17$~dB; phase preservation adds $+1.03$~dB. A novel spatial shuffling ablation ($-1.26$~dB penalty) demonstrates phase encodes location-dependent structure. We conduct a preliminary extensibility study on a second dense prediction task (ISIC skin lesion segmentation), with full cross-validation as ongoing work. This work advances principled wavelet-deep learning integration, showing how phase information complements scattering's stability-expressiveness trade-off in pixel-level prediction.
format Preprint
id arxiv_https___arxiv_org_abs_2605_24621
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Phase-Aware Wavelet-Based-Scattering Encoder-Decoder for Dense Predictions
Marrakchi, Ghassen
Matei, Basarab
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
I.2.10; I.4.4; I.4.6; I.5.1; I.5.2
Scattering transforms achieve Lipschitz stability and translation invariance, but dense prediction tasks require preserving spatial structure lost in global averaging. We propose Phase-Aware Scattering Encoder-Decoder, which restores this information by explicitly preserving phase in skip connections. On image denoising (BSD68), breaking translation invariance improves PSNR by $+2.17$~dB; phase preservation adds $+1.03$~dB. A novel spatial shuffling ablation ($-1.26$~dB penalty) demonstrates phase encodes location-dependent structure. We conduct a preliminary extensibility study on a second dense prediction task (ISIC skin lesion segmentation), with full cross-validation as ongoing work. This work advances principled wavelet-deep learning integration, showing how phase information complements scattering's stability-expressiveness trade-off in pixel-level prediction.
title Phase-Aware Wavelet-Based-Scattering Encoder-Decoder for Dense Predictions
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
I.2.10; I.4.4; I.4.6; I.5.1; I.5.2
url https://arxiv.org/abs/2605.24621