AI Paper Atlas

Research map / 3D vision / Depth and stereo

Surface Normals and Shape From Water

Satoshi Murai, Meng-Yu Jennifer Kuo, Ryo Kawahara, Shohei Nobuhara, Ko Nishino

ICCV 2019 · 3D vision › Depth and stereo · approach: Non-line-of-sight imaging · 5 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 per-pixel simultaneous normal and depth estimation instead of deriving normals from depthSurface normals and shape from water · Synthetic data, K=4 (object 1)normal RMSE (degrees): 29.054 -> 0.069Improved
add per-pixel simultaneous normal and depth estimation instead of integrating depth from normalsSurface normals and shape from water · Synthetic data, K=4 (object 1)depth RMSE: 0.030 -> 0.007Improved
add direct normal estimation instead of normals from depthSurface normals and shape from water · Synthetic data, K=4 (object 2)normal RMSE (degrees): 51.674 -> 1.619Improved
add direct depth estimation instead of integrating normalsSurface normals and shape from water · Synthetic data, K=4 (object 2)depth RMSE: 0.053 -> 0.015Improved
add light direction and intensity calibration with an immersed sphere (ablation)Surface normals and shape from water (real system) · Real calibration spherenormal error En: 27.039 -> 7.728Improved
add light direction and intensity calibration with an immersed sphere (ablation)Surface normals and shape from water (real system) · Real calibration spheredepth error Ed: 0.183 -> 0.002Improved
increase the number of auxiliary light sources from 3 to 8 (ablation)Surface normals and shape from water · Synthetic data with Gaussian noisenormal and depth errorImproved
use multi-directional multi-wavelength lights instead of two coaligned lights (bispectral shape from water)Surface normals and shape from water · Synthetic data with Gaussian noisedepth and normal errorImproved

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