AI Paper Atlas

Research map / 3D vision / 3D generation

Semantic Visual Localization

Johannes L. Schönberger, Marc Pollefeys, Andreas Geiger, Torsten Sattler

CVPR 2018 · 3D vision › 3D generation · approach: Semantic scene completion · 286 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
Use generative semantic scene-completion descriptors instead of the evaluated geometric shape descriptorsSemantic Visual Localization · KITTI 90° viewpoint localizationlocalization performanceImproved
Add semantic information to the learned descriptors (ablation)Semantic Visual Localization · KITTI 90° viewpoint localizationlocalization performanceImproved
Accumulate evidence from five consecutive query frames (ablation)Semantic Visual Localization · KITTI 90° viewpoint localizationlocalization performanceImproved
Use the semantic descriptor instead of the geometric-only descriptor (ablation)Semantic Visual Localization · KITTI 0° viewpoint localizationlocalization performanceNo clear effect
Use the semantic descriptor instead of the geometric-only descriptor (ablation)Semantic Visual Localization · NCLT cross-time localizationlocalization performanceImproved
Use a latent code size below 64 (ablation)Semantic Visual Localization · Descriptor learningreconstruction and localization performanceHurt
Increase latent code size above 512 (ablation)Semantic Visual Localization · Descriptor learningreconstruction and localization performanceNo clear effect
Use SIFT instead of semantic 3D descriptorsVisual localization pipeline · KITTI 90° viewpoint localizationlocalization performanceHurt
Use multiple accumulated query frames rather than a single-frame query (ablation)Semantic Visual Localization · Localization across viewpoint and appearance changeslocalization performanceImproved

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