Spatial Frequency Modulation for Semantic Segmentation

TL;DR

This paper introduces Spatial Frequency Modulation (SFM), a novel method that reduces aliasing of high-frequency details during downsampling in segmentation models, improving accuracy across multiple vision tasks.

Contribution

The paper proposes a new SFM technique with adaptive resampling and multi-scale upsampling modules that effectively preserve high-frequency information in segmentation models.

Findings

SFM reduces aliasing and retains details in segmentation.

SFM improves performance across various vision tasks.

The method is compatible with different neural network architectures.

Abstract

High spatial frequency information, including fine details like textures, significantly contributes to the accuracy of semantic segmentation. However, according to the Nyquist-Shannon Sampling Theorem, high-frequency components are vulnerable to aliasing or distortion when propagating through downsampling layers such as strided-convolution. Here, we propose a novel Spatial Frequency Modulation (SFM) that modulates high-frequency features to a lower frequency before downsampling and then demodulates them back during upsampling. Specifically, we implement modulation through adaptive resampling (ARS) and design a lightweight add-on that can densely sample the high-frequency areas to scale up the signal, thereby lowering its frequency in accordance with the Frequency Scaling Property. We also propose Multi-Scale Adaptive Upsampling (MSAU) to demodulate the modulated feature and recover high-frequency information through non-uniform upsampling This module further improves segmentation by explicitly exploiting information interaction between densely and sparsely resampled areas at multiple scales. Both modules can seamlessly integrate with various architectures, extending from convolutional neural networks to transformers. Feature visualization and analysis confirm that our method effectively alleviates aliasing while successfully retaining details after demodulation. Finally, we validate the broad applicability and effectiveness of SFM by extending it to image classification, adversarial robustness, instance segmentation, and panoptic segmentation tasks. The code is available at https://github.com/Linwei-Chen/SFM.