AI Paper Atlas

Research map / 3D vision / Depth and stereo

Structural Deep Metric Learning for Room Layout Estimation

Wenzhao Zheng, Jiwen Lu, Jie Zhou

ECCV 2020 · 3D vision › Depth and stereo · approach: Foundation models for monocular geometry · 21 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 Lmetric (structural relation loss) (ablation)SDML · LSUNpixel error (%): 9.93 -> 6.95Improved
add Lmetric (structural relation loss) (ablation)SDML · LSUNkeypoint error (%): 6.42 -> 5.29Improved
add Lstr (structural prediction loss) (ablation)SDML · LSUNpixel error (%): 15.36 -> 6.95Improved
add Lstr (structural prediction loss) (ablation)SDML · LSUNkeypoint error (%): 10.23 -> 5.29Improved
use Lmetric instead of dense structural loss Ldense (ablation)SDML · LSUNpixel error (%): 8.56 -> 6.95Improved
use Lmetric instead of dense structural loss Ldense (ablation)SDML · LSUNkeypoint error (%): 5.86 -> 5.29Improved
add dense structural loss to RoomNet type classifier (ablation)RoomNet · LSUNpixel error (%): 10.46 -> 9.23Improved
add dense structural loss to RoomNet type classifier (ablation)RoomNet · LSUNkeypoint error (%): 6.95 -> 6.27Improved
add layout decoder (replace keypoint regression) (ablation)SDML · LSUNpixel error (%): 9.97 -> 6.95Improved
add layout decoder (replace keypoint regression) (ablation)SDML · LSUNkeypoint error (%): 6.60 -> 5.29Improved
replace RoomNet L2 keypoint regression with keypoint regression without decoder in SDML framework (ablation)keypoint regression · LSUNpixel error (%): 10.46 -> 9.97Improved
use dense structural loss instead of log-ratio dense samplingmetric model · LSUN layout retrieval (1 retrieval)mean label distance (%): 23.3 -> 20.7Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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