CVOct 11, 2022

Deep Fourier Up-Sampling

arXiv:2210.05171v174 citationsh-index: 128
Originality Incremental advance
AI Analysis

This work addresses the problem of global dependency in up-sampling for computer vision researchers, offering a novel operator that can be integrated into existing networks, though it appears incremental as it builds on known Fourier principles.

The paper tackles the limitation of spatial up-sampling in convolutional neural networks by proposing Deep Fourier Up-Sampling (FourierUp), a method that leverages the Fourier domain for global modeling, resulting in consistent performance gains across multiple computer vision tasks such as object detection and image segmentation.

Existing convolutional neural networks widely adopt spatial down-/up-sampling for multi-scale modeling. However, spatial up-sampling operators (\emph{e.g.}, interpolation, transposed convolution, and un-pooling) heavily depend on local pixel attention, incapably exploring the global dependency. In contrast, the Fourier domain obeys the nature of global modeling according to the spectral convolution theorem. Unlike the spatial domain that performs up-sampling with the property of local similarity, up-sampling in the Fourier domain is more challenging as it does not follow such a local property. In this study, we propose a theoretically sound Deep Fourier Up-Sampling (FourierUp) to solve these issues. We revisit the relationships between spatial and Fourier domains and reveal the transform rules on the features of different resolutions in the Fourier domain, which provide key insights for FourierUp's designs. FourierUp as a generic operator consists of three key components: 2D discrete Fourier transform, Fourier dimension increase rules, and 2D inverse Fourier transform, which can be directly integrated with existing networks. Extensive experiments across multiple computer vision tasks, including object detection, image segmentation, image de-raining, image dehazing, and guided image super-resolution, demonstrate the consistent performance gains obtained by introducing our FourierUp.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes