AI Paper Atlas

Research map / 3D vision / Depth and stereo

Improving Monocular Depth Estimation by Leveraging Structural Awareness and Complementary Datasets

Tian Chen, Shijie An, Yuan Zhang, Chongyang Ma, Huayan Wang, Xiaoyan Guo, Wen Zheng

ECCV 2020 · 3D vision › Depth and stereo · approach: Self-supervised monocular depth · 31 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 edge-aware consistency (ablation)baseline Mb · NYUDv2RMSE: 0.430 -> 0.426Improved
add spatial attention blocks (SAB) (ablation)Me · NYUDv2RMSE: 0.426 -> 0.417Improved
add global focal relative loss (GFRL) (ablation)Me,SAB · NYUDv2RMSE: 0.417 -> 0.407Improved
use GFRL instead of original relative loss (ablation)Me,SAB · NYUDv2delta1 accuracy: 0.864 -> 0.876Improved
use SAB instead of sSE (ablation)Me,Lr · NYUDv2RMSE: 0.433 -> 0.407Improved
use SAB instead of CBAM (ablation)Me,Lr · NYUDv2RMSE: 0.432 -> 0.407Improved
add HC Depth dataset to training (ablation)Mfull · NYUDv2delta1 accuracy: 0.876 -> 0.885Improved
train on all six datasets (ablation)Mfull · NYUDv2RMSE: 0.407 -> 0.376Improved
train on all six datasets (ablation)Mfull · TUM (zero-shot)RMSE: 1.109 -> 0.912Improved
add HC Depth dataset only (ablation)Mfull · HC Depth testdelta1 accuracy: 0.600 -> 0.825Improved
train on all six datasets (ablation)Mfull · HC Depth testdelta1 accuracy: 0.600 -> 0.879Improved
add synthetic and indoor datasets 2,3 only (ablation)Mfull · HC Depth testdelta1 accuracy: 0.600 -> 0.610Improved

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