AI Paper Atlas

Research map / 3D vision / Depth and stereo

Virtual Normal: Enforcing Geometric Constraints for Accurate and Robust Depth Prediction

Wei Yin, Yifan Liu, Chunhua Shen

TPAMI 2021 · 3D vision › Depth and stereo · approach: Self-supervised monocular depth · 82 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 virtual normal loss to WCEL (ablation)WCEL baseline · NYUD-V2Abs-Rel: 0.1427 -> 0.1337Improved
add L1 loss to WCEL (ablation)WCEL baseline · NYUD-V2Abs-Rel: 0.1427 -> 0.1429No clear effect
add surface normal loss (local) to WCEL (ablation)WCEL baseline · NYUD-V2Abs-Rel: 0.1427 -> 0.1406Improved
add pair-wise geometric loss to WCEL (ablation)WCEL baseline · NYUD-V2Abs-Rel: 0.1427 -> 0.1380Improved
add virtual normal loss on top of WCEL+PL (ablation)WCEL+PL · NYUD-V2Abs-Rel: 0.1380 -> 0.1341Improved
increase virtual normal sample groups from 0K to 20K (ablation)WCEL+VNL · NYUD-V2Abs-Rel: 5.6%Improved
use multi-curriculum learning instead of uniform sampling (ablation)DiverseDepth training · KITTI zero-shotAbs-Rel: 17.9 -> 12.6Improved
reverse curriculum (hard to easy) instead of uniform sampling (ablation)DiverseDepth training · KITTI zero-shotAbs-Rel: 17.9 -> 15.8Improved
use easy-to-hard curriculum instead of reverse curriculum (ablation)DiverseDepth training · KITTI zero-shotAbs-Rel: 15.8 -> 12.6Improved
add SSIL to VNL (ablation)affine-invariant depth model · KITTI zero-shotAbs-Rel: 21.0 -> 12.6Improved
add VNL to SSIL (ablation)affine-invariant depth model · KITTI zero-shotAbs-Rel: 16.3 -> 12.6Improved
replace MSE loss with VNL+SSIL (ablation)affine-invariant depth model · KITTI zero-shotAbs-Rel: 36.0 -> 12.6Improved

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