AI Paper Atlas

Research map / Image generation / Image restoration

Stereo Computation for a Single Mixture Image

Yiran Zhong, Yuchao Dai, Hongdong Li

ECCV 2018 · Image generation › Image restoration · approach: Inverse imaging with learned priors · 15 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
add stereo computation (matching) loss jointly with image separation (ablation)Image separation module · KITTI image separationPSNR: 19.5009 -> 20.0914Improved
use sub-pixel disparity optimization instead of pixel-level disparity (vs Williem et al.)De-anaglyph · Middlebury de-anaglyphbad pixel ratio (%): 7.86 -> 5.93Improved
include disparity in the optimization (ours vs Joulin et al. which does not)De-anaglyph · Middlebury de-anaglyphbad pixel ratio (%): 26.82 -> 5.93Improved
learned joint model vs Williem et al. iterative methodDe-anaglyph · KITTI 2015 D1_allbad pixel ratio (%): 14.40% -> 5.96%Improved
learned joint model vs layer separation method on double-vision imagesDiplopia unmixing · KITTI 2015 double vision (right view)PSNR: 7.67 -> 24.55Improved
use estimated mixture-based disparity vs oracle disparity from clean stereoDiplopia unmixing · KITTI 2015 double visionmean bad pixel ratio (%): 5.28% -> 6.67%Mixed
apply general mixture framework to monocular depth vs Godard et al.Monocular depth estimation · KITTI 2015 monocular depth (Eigen split)Abs Rel: 0.140 -> 0.126Improved
apply general mixture framework to monocular depth vs Godard et al.Monocular depth estimation · KITTI 2015 monocular depth (Eigen split)RMSE: 4.471 -> 3.971Improved

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