AI Paper Atlas

Research map / 3D vision / Depth and stereo

FourierNets enable the design of highly non-local optical encoders for computational imaging

Diptodip Deb, Zhenfei Jiao, Ruth R Sims, Alex Bo-Yuan Chen, Michael Broxton, Misha Ahrens, Kaspar Podgorski, Srinivas C Turaga

NeurIPS 2022 · 3D vision › Depth and stereo · approach: Neural light field and illumination imaging · 10 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use global Fourier convolutions rather than the UNet decoderDiffuserCam lensless photography reconstruction · DiffuserCam Lensless Mirflickr Dataset testMSE (×10^-2): 1.68 ± 0.060 -> 0.39 ± 0.007Improved
Use global Fourier convolutions rather than Le-ADMM-UDiffuserCam lensless photography reconstruction · DiffuserCam Lensless Mirflickr Dataset testMSE (×10^-2): 0.75 ± 0.021 -> 0.39 ± 0.007Improved
Train FourierNet with MSE+LPIPS instead of MSE loss (ablation)FourierNet · DiffuserCam Lensless Mirflickr Dataset testLPIPS: 0.20 ± 0.00 -> 0.16 ± 0.00Improved
Train FourierNet with MSE+LPIPS instead of MSE loss (ablation)FourierNet · DiffuserCam Lensless Mirflickr Dataset testMSE (×10^-2): 0.39 ± 0.007 -> 0.54 ± 0.010Hurt
Use FourierNet2D rather than UNet2D to optimize the microscope encodingZebrafish snapshot microscopy · Zebrafish snapshot microscopy, 256×256 cameraL_HNMSE: 0.7109 ± 0.0161 -> 0.6093 ± 0.0209Improved
Use FourierNet3D rather than UNet3D for reconstructionFourierNet2D-optimized microscope · Zebrafish snapshot microscopy, 256×256 cameraL_HNMSE: 0.7298 ± 0.0151 -> 0.6093 ± 0.0209Improved
Add Wiener deconvolution before the UNetUNet-based microscopy decoder · Zebrafish snapshot microscopy, 256×256 cameraL_HNMSE: 0.7109 ± 0.0161 -> 0.7223 ± 0.0179Mixed
Add multiscale Fourier features to FourierNet (ablation)FourierNet lensless reconstruction · DiffuserCam Lensless Mirflickr Dataset testMSE (×10^-2): 0.39 ± 0.007 -> 0.43 ± 0.009Hurt
Add multiscale Fourier features to FourierNet (ablation)FourierNet 3D volume reconstruction · Zebrafish snapshot microscopy, 256×256 cameraPSNR: 34.89 ± 0.88 -> 34.87 ± 0.82No clear effect
Optimize microscope parameters for the same ROI type used at test timeProgrammable zebrafish snapshot microscope · Type A ROIPSNR: 46.03 ± 1.33 -> 49.76 ± 1.35Improved
Optimize microscope parameters for the same ROI type used at test timeProgrammable zebrafish snapshot microscope · Type B ROIPSNR: 35.56 ± 1.41 -> 37.34 ± 0.96Improved
Optimize microscope parameters for the same ROI type used at test timeProgrammable zebrafish snapshot microscope · Type C ROIPSNR: 30.87 ± 1.15 -> 33.79 ± 0.90Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy