Fourier transformers for super-resolution images and CT scans

Fourier transformers for super-resolution images and CT scans

Fourier Image Transformer

arXiv paper abstract https://arxiv.org/abs/2104.02555

arXiv PDF paper https://arxiv.org/pdf/2104.02555.pdf

GitHub https://github.com/juglab/FourierImageTransformer

We present Fourier Image Transformers (FITs), realizations of Fast-Transformers (7), that operate on images via a novel sequential image representation we call Fourier Domain Encoding (FDE). We demonstrate the utility of FITs on two tasks. First, an image super-resolution task (top row), where a given low-resolution image is used to predict a higher resolution version. The second task is tomographic reconstruction (bottom row), where a given set of projection images (a sinogram) is transformed by an encoder-decoder FIT to improve the quality of the reconstructed image w.r.t. filtered backprojection (FBP), a commonly used method for tomographic reconstructions. Note that Fourier Image Transformers solve both tasks in Fourier space, with the final results being obtained by an inverse Fourier transformation (iFFT).

Please like and share this post if you enjoyed it using the buttons at the bottom!

Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact

Web site with my other posts by category https://morrislee1234.wixsite.com/website

#ComputerVision #FFT #SuperResolution #CT #AINewsClips #AI #ML #ArtificialIntelligence #MachineLearning