Phase-Aware Wavelet-Based-Scattering Encoder-Decoder for Dense Predictions
For researchers in image denoising and dense prediction, this work shows how phase information complements scattering transforms, but the results are preliminary with only one task fully evaluated.
The paper proposes a Phase-Aware Scattering Encoder-Decoder that preserves phase in skip connections to restore spatial structure lost in global averaging. On BSD68 image denoising, breaking translation invariance improves PSNR by +2.17 dB, and phase preservation adds +1.03 dB.
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.