AI Paper Atlas

Research map / 3D vision / Depth and stereo

A Generative Model for Generic Light Field Reconstruction

Paramanand Chandramouli, Kanchana Vaishnavi Gandikota, Andreas Görlitz, Andreas Kolb, Michael Moeller

TPAMI 2020 · 3D vision › Depth and stereo · approach: Neural light field and illumination imaging · 16 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
average over overlapping patches with stride 5 (ablation)CVAE prior · 3x3 to 7x7 view synthesisPSNR (dB): 38.53 -> 39.77Improved
average over overlapping patches with stride 5 (ablation)CVAE prior · 5x5 view synthesis, mask M1, DinoPSNR (dB): 39.57 -> 41.53Improved
average over overlapping patches with stride 5 (ablation)CVAE prior · 5x5 coded aperture, mask M1, DinoPSNR (dB): 34.97 -> 38.46Improved
use CVAE prior with energy minimization instead of dictionary priorenergy-minimization LF reconstruction · 5x5 view synthesis, mask M1PSNR (dB): by 2.63 dB (mask M1), 3.13 dB (mask M2)Improved
use CVAE prior with overlapping patches instead of dictionary priorenergy-minimization LF reconstruction · 5x5 view synthesisPSNR (dB): by 4 dB and 4.4 dBImproved
reduce known views from 3x3 to 5 randomly chosen views (ablation)CVAE prior with overlapping patches · 7x7 view synthesisPSNR (dB): 39.77 -> 39.57Hurt
use CVAE prior instead of end-to-end CNN with noisy inputs (sigma 0.1)CVAE prior · Cars 3x3 to 7x7, Gaussian noise 0.1PSNR (dB): 29.95 vs 31.67Improved
use CVAE prior instead of end-to-end CNN with 50% pixel dropCVAE prior · Cars 3x3 to 7x7, 50% pixel dropPSNR (dB): 13.60 vs 31.68Improved
use L1 loss instead of MSE loss for salt and pepper noise (ablation)CVAE prior · Cars, salt and pepper noisePSNR (dB): PSNR dropped by about 2 dB with MSE vs 0.1 dB with L1Improved
alter coded mask with swaps at test time (1 swap)end-to-end network · Dino coded aperture M1PSNR (dB): 38.7 -> 24.3Hurt
alter coded mask with swaps at test time (3 swaps)CVAE prior with overlapping patches · Dino coded aperture M1PSNR (dB): 38.46 -> 38.19No clear effect
add TV penalty on central view when all views are corruptedCVAE prior · Cars, all views corruptedPSNR (dB)Improved

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