AI Paper Atlas

Research map / 3D vision / Depth and stereo

Domain Randomization-Enhanced Depth Simulation and Restoration for Perceiving and Grasping Specular and Transparent Objects

Qiyu Dai, Jiyao Zhang, Qiwei Li, Tianhao Wu, Hao Dong, Ziyuan Liu, Ping Tan, He Wang

ECCV 2022 · 3D vision › Depth and stereo · approach: Depth from light field and defocus · 83 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
train on simulated DREDS depth instead of cropped synthetic depth (ablation)SwinDRNet · STD-CatKnownRMSE: 0.0467 -> 0.015Improved
train on simulated DREDS depth instead of cropped synthetic depth (ablation)SwinDRNet · STD-CatNovelRMSE: 0.065 -> 0.025Improved
train on DREDS-CatKnown and transfer to ClearGrasp instead of training on ClearGrasp and OmniverseSwinDRNet · ClearGrasp real-knownRMSE: 0.028 -> 0.022Improved
train on DREDS-CatKnown and transfer to ClearGrasp instead of training on ClearGrasp and OmniverseSwinDRNet · ClearGrasp real-novelRMSE: 0.025 -> 0.016Improved
add RGB-D fusion module (ablation)SwinDRNet · STD-CatKnownRMSE: 0.019 -> 0.014Improved
add confidence interpolation module (ablation)SwinDRNet · STD-CatKnownRMSE: 0.014 -> 0.015No clear effect
add confidence interpolation module (ablation)SwinDRNet+NOCSHead · STD-CatKnown5deg 5cm pose accuracy: 52.2 -> 54.1Improved
add material randomization instead of fixed ShapeNet materials (ablation)SwinDRNet · STD-CatKnownRMSE: 0.024 -> 0.015Improved
add material randomization instead of fixed ShapeNet materials (ablation)SwinDRNet · STD-CatNovelRMSE: 0.038 -> 0.025Improved
double training data scale from half to full DREDS (ablation)SwinDRNet · STD-CatKnownRMSE: 0.021 -> 0.018Improved
double training data scale from half to full DREDS (ablation)SwinDRNet · STD-CatNovelRMSE: 0.028 -> 0.025Improved
use restored depth from SwinDRNet for pose fitting instead of raw depth (ablation)SwinDRNet+NOCSHead · STD-CatKnown5deg 5cm pose accuracy: 42.9 -> 54.1Improved

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