AI Paper Atlas

Research map / 3D vision / Depth and stereo

Pixel-Pair Occlusion Relationship Map (P2ORM): Formulation, Inference & Application

Xuchong Qiu, Yang Xiao, Chaohui Wang, Renaud Marlet

ECCV 2020 · 3D vision › Depth and stereo · approach: Self-supervised monocular depth

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
use pixel-pair occlusion relationship (4-connectivity) instead of separate boundary and orientation headsResNet U-Net · BSDS ownershipODS: .571 -> .590Improved
use pixel-pair occlusion relationship (4-connectivity) instead of separate boundary and orientation headsResNet U-Net · NYUv2-ORODS: .396 -> .500Improved
use pixel-pair occlusion relationship (4-connectivity) instead of separate boundary and orientation headsResNet U-Net · iBims-1-ORODS: .482 -> .575Improved
add diagonal neighbors (8-connectivity instead of 4) (ablation)P2ORM · BSDS ownershipODS: .590 -> .607Improved
add diagonal neighbors (8-connectivity instead of 4) (ablation)P2ORM · NYUv2-ORODS: .500 -> .520Improved
add diagonal neighbors (8-connectivity instead of 4) (ablation)P2ORM · iBims-1-ORODS: .575 -> .581Improved
add diagonal neighbors (8-connectivity instead of 4) (ablation)P2ORM · NYUv2-ORAP: .477 -> .497Improved
use plain ResNet U-Net baseline instead of specialized occlusion architecturesbaseline variant · BSDS ownershipODS: .553 -> .571No clear effect
refine SharpNet depth with estimated P2ORMdepth refinement · NYUv2 and iBims-1depth-edge accuracy and completionImproved
refine depth with P2ORMdepth refinement · NYUv2 and iBims-1general depth metrics (rel, RMSE): <1%No clear effect
use the proposed refinement variant over alternatives (extra inputs, GT-depth loss, d_pq only, D_pq only) (ablation)depth refinement · iBims-1 / NYUv2depth-edge metricsImproved

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