AI Paper Atlas

Research map / 3D vision / 3D generation

3D Scene Graph: A Structure for Unified Semantics, 3D Space, and Camera

Iro Armeni, Zhi-Yang He, JunYoung Gwak, Amir R. Zamir, Martin Fischer, Jitendra Malik, Silvio Savarese

ICCV 2019 · 3D vision › 3D generation · approach: Open-vocabulary 3D scene understanding · 399 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 framing of query images sampled on panoramas (ablation)Mask R-CNN · Gibson 2D panoramasAP: 0.079 -> 0.160Improved
add multi-view consistency on top of framing (ablation)Mask R-CNN + Framing · Gibson 2D panoramasAP: 0.160 -> 0.485Improved
add framing before panorama projection (ablation)Mask R-CNN + Pano Projection · Gibson 3D meshAP: 0.222 -> 0.306Improved
add multi-view consistency on top of framing (ablation)Mask R-CNN + Framing + Pano Projection · Gibson 3D meshAP: 0.306 -> 0.409Improved
add framing to a weaker detector (ablation)BlitzNet · Gibson 2D panoramasAP: 0.095 -> 0.198Improved
add multi-view consistency to a weaker detector (ablation)BlitzNet + Framing · Gibson 2D panoramasAP: 0.198 -> 0.284Improved
add framing and MVC to a weaker detector (ablation)BlitzNet · Gibson 3D meshAP: 0.076 -> 0.245Improved
add human manual verification to the automatic pipelineOurs (fully automatic) · Gibson tiny splitAP: 0.389 -> 0.97Improved
use automatic pipeline plus verification instead of fully manual 2D annotationOurs w/ human (MV) · Gibson annotation timehours: 12:44:10 -> 03:18:02Improved
train on 3D-scene-graph-derived masks for spatial orderResNet34 · Spatial order classificationmean AP: 0.255 -> 0.712Improved
train on 3D-scene-graph-derived masks for relative volumeResNet34 · Relative volume classificationmean AP: 0.555 -> 0.820Improved
learn amodal mask prediction with U-Net instead of class-specific average maskU-Net · Amodal mask segmentationavg f1-score: 0.545 -> 0.672Improved

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