Scaling Up Computer Vision Neural Networks Using Fast Fourier Transform

Siddharth Agrawal
2/2/2023
cs.CVcs.LG

No official code found. Best implementation: siddagra/fourier-project (5 stars, last commit 2025-06-13).

Abstract

Deep Learning-based Computer Vision field has recently been trying to explore larger kernels for convolution to effectively scale up Convolutional Neural Networks. Simultaneously, new paradigm of models such as Vision Transformers find it difficult to scale up to larger higher resolution images due to their quadratic complexity in terms of input sequence. In this report, Fast Fourier Transform is utilised in various ways to provide some solutions to these issues.

View on arXivView PDF

Code Implementations(1)

siddagra/fourier-projectCommunity100%
50PythonNov 14, 20221 years ago

Similar Papers

Beyond Uniform Scaling: Exploring Depth Heterogeneity in Neural Architectures

82% match2024

Accelerating Vision Transformers on Brain Processing Unit

81% match2026

ScaleNet: Scaling up Pretrained Neural Networks with Incremental Parameters

80% match1 repo32025

ConvNets Match Vision Transformers at Scale

80% match1 repo62023

Scaled Quantization for the Vision Transformer

80% match2023

Multiscale Attention via Wavelet Neural Operators for Vision Transformers

80% match2023

Cite this paper

@article{agrawal2023scaling,
  title  = {Scaling Up Computer Vision Neural Networks Using Fast Fourier Transform},
  author = {Siddharth Agrawal},
  year   = {2023},
  eprint = {2302.12185},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2302.12185v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2302.12185.svg)](https://codeofpaper.com/paper/2302.12185)

Discussion