AI Paper Atlas

Research map / 3D vision / Depth and stereo

Image-Based Localization Using LSTMs for Structured Feature Correlation

Florian Walch, Caner Hazirbas, Laura Leal-Taixe, Torsten Sattler, Sebastian Hilsenbeck, Daniel Cremers

ICCV 2017 · 3D vision › Depth and stereo · approach: Scene coordinate regression for localization · 525 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 LSTM-based structured dimensionality reduction after the FC layer (vs PoseNet)PoseNet (GoogLeNet) · Cambridge Landmarksmedian position error (m): 2.08 -> 1.30Improved
add LSTM-based structured dimensionality reduction (vs PoseNet)PoseNet (GoogLeNet) · Cambridge Landmarksmedian orientation error (deg): 6.83 -> 5.52Improved
add LSTM structured reduction instead of dropout (vs Bayesian PoseNet)Bayesian PoseNet · Cambridge Landmarksmedian position error (m): 1.92 -> 1.30Improved
add LSTM structured reduction (vs PoseNet)PoseNet (GoogLeNet) · 7Scenesmedian position error (m): 0.44 -> 0.31Improved
add LSTM structured reduction (vs PoseNet)PoseNet (GoogLeNet) · 7Scenesmedian orientation error (deg): 10.4 -> 9.85Improved
add LSTM structured reduction (vs PoseNet)PoseNet (GoogLeNet) · 7Scenes - Headsmedian orientation error (deg): 12.0 -> 13.7Hurt
add LSTM structured reduction (vs PoseNet)PoseNet (GoogLeNet) · 7Scenes - Pumpkinmedian position error (m): 0.61 -> 0.33Improved
add LSTM structured reduction (vs PoseNet)PoseNet (GoogLeNet) · TUM-LSImedian position error (m): 1.87 -> 1.31Improved
add LSTM structured reduction (vs PoseNet)PoseNet (GoogLeNet) · TUM-LSImedian orientation error (deg): 6.14 -> 2.79Improved
use CNN pose regression instead of SIFT-based Active SearchActive Search (w/) · Cambridge Landmarksmedian position error (m): 0.36 -> 1.37Hurt
use CNN pose regression instead of SIFT-based Active SearchActive Search (w/) · 7Scenesmedian position error (m): 0.06 -> 0.30Hurt
feed reshaped 32x64 matrix to four directional LSTMs instead of the raw 2048-D vector (ablation)CNN+LSTM · Cambridge/7Scenespose accuracyImproved

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