AI Paper Atlas

Research map / 3D vision / Human pose estimation

Cascaded Parallel Filtering for Memory-Efficient Image-Based Localization

Wentao Cheng, Weisi Lin, Kan Chen, Xinfeng Zhang

ICCV 2019 · 3D vision › Human pose estimation · approach: Learned camera pose estimation · 32 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 bilateral Hamming ratio test with two-step match selection (vs baseline voting) (ablation)Our method · Dubrovniklocalized images: 778 -> 794Improved
add bilateral Hamming ratio test with two-step match selection (vs baseline voting) (ablation)Our method · Dubrovnikmedian error (m): 0.69 -> 0.64Improved
add quality-aware spatial reconfiguration (QSR) (ablation)Our method · Dubrovnikmedian error (m): 0.74 -> 0.64Improved
add quality-aware spatial reconfiguration (QSR) (ablation)Our method · Dubrovnik75% error quartile (m): 2.53 -> 2.16Improved
add quality-aware spatial reconfiguration (QSR) (ablation)Our method · Dubrovniklocalized images: 793 -> 794No clear effect
add principal focal length estimation (PFL) (ablation)Our method · Dubrovnikmedian error (m): 0.80 -> 0.64Improved
add principal focal length estimation (PFL) (ablation)Our method · Dubrovnik75% error quartile (m): 2.70 -> 2.16Improved
store compact binary signatures instead of integer mean SIFT descriptors (vs Active Search)Our method · Dubrovnikmemory (GB): 0.75 -> 0.14Improved
store compact binary signatures instead of integer mean SIFT descriptors (vs Active Search)Our method · RobotCar Seasonsmemory (GB): 2.72 -> 0.52Improved
store compact binary signatures instead of integer mean SIFT descriptors (vs Active Search)Our method · Aachen Day-Nightmemory (GB): 0.76 -> 0.14Improved
use cascaded filtering without gravity prior vs CSL with gravity priorOur method · Aachen Day-Night (day, 0.25m/2deg)localized %: 52.3 -> 76.7Improved
use no semantic network vs SMC with semantic segmentationOur method · RobotCar Seasons (all night, 0.25m/2deg)localized %: 7.6 -> 3.4Hurt

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