AI Paper Atlas

Research map / 3D vision / Depth and stereo

Into the Twilight Zone: Depth Estimation using Joint Structure-Stereo Optimization

Aashish Sharma, Loong-Fah Cheong

ECCV 2018 · 3D vision › Depth and stereo · approach: Depth from light field and defocus · 9 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
replace DnCNN denoising + MeshStereo with joint structure-stereo optimizationDnCNN+MS · Middlebury v3, noise 60, 1pxbad pixel percentage: 58.97 -> 56.59Improved
replace BM3D denoising + MeshStereo with joint structure-stereo optimizationBM3D+MS · Middlebury v3, noise 60, 1pxbad pixel percentage: 63.35 -> 56.59Improved
replace SS-PCA joint denoising with joint structure-stereo optimizationSS-PCA · Middlebury v3, noise 60, 1pxbad pixel percentage: 69.17 -> 56.59Improved
use joint structure-stereo optimization at low noise instead of denoising baselineDnCNN+MS · Middlebury v3, noise 25, 1pxbad pixel percentage: 43.43 -> 47.74Hurt
use joint structure-stereo optimization at low noise instead of denoising baselineDnCNN+MS · Middlebury v3, noise 25, 3pxbad pixel percentage: 22.04 -> 25.12Hurt
use joint structure-stereo optimization at medium noiseDnCNN+MS · Middlebury v3, noise 50, 1pxbad pixel percentage: 55.06 -> 54.19Improved
use joint structure-stereo optimization at noise 60, 3pxDnCNN+MS · Middlebury v3, noise 60, 3pxbad pixel percentage: 32.68 -> 30.48Improved
use joint structure-stereo optimization on poorly-lit real night imagesDnCNN+MS · Oxford RobotCar Set1, 5pxbad pixel percentage: 19.66 -> 12.31Improved
use joint structure-stereo optimization on poorly-lit real night imagesDnCNN+MS · Oxford RobotCar Set1, 1pxbad pixel percentage: 63.86 -> 58.76Improved
use joint structure-stereo optimization on well-lit real night imagesDnCNN+MS · Oxford RobotCar Set2, 2pxbad pixel percentage: 28.82 -> 28.80No clear effect
fine tune parameters on Set2Joint Structure-Stereo · Oxford RobotCar Set2, 1pxbad pixel percentage: 57.76 -> 56.45Improved
fine tune parameters on Set2Joint Structure-Stereo · Oxford RobotCar Set2, 5pxbad pixel percentage: 6.82 -> 6.08Improved

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