AI Paper Atlas

Research map / 3D vision / Depth and stereo

Global Underwater Geolocation from Time-Lapse Polarization Imagery

Sara Aghajanzadeh, Xiaoyang Bai, Zhongmin Zhu, David Forsyth, Viktor Gruev

CVPR 2026 · 3D vision › Depth and stereo · approach: Non-line-of-sight imaging

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 day-of-year token (ablation)Polar Transformer · Cross-site, ChampaignRMSE (degrees): 6.5 -> 5.2Improved
use absolute instead of relative attention in Point Transformer (ablation)Polar Transformer · Cross-site, Champaignmedian geodesic error (km): 845 km -> 567 kmImproved
add trajectory-dropout regularization (ablation)Polar Transformer · Cross-site, Champaignmedian geodesic error (km): 567 km -> 466 kmImproved
double sequence length from 64 to 128 (ablation)Polar Transformer · Cross-site, ChampaignRMSE (degrees): RMSE 3° with marginal geolocation gainsNo clear effect
add ViT encoder summary token (ablation)Polar Transformer · Cross-site, ChampaignRMSE and outliersImproved
use Point Transformer instead of MLP for geolocation head (ablation)Polar Transformer · Cross-site, Champaignmedian geodesic error (km): above 620 km -> 466 kmImproved
train on simulated sequences excluding target-site frames instead of real target-site data (ablation)Polar Transformer · Same-site geolocationgeodesic errorNo clear effect
include target-site frames in simulator synthesis (variant b vs c) (ablation)Polar Transformer · Same-site geolocationgeodesic errorNo clear effect
train on synthetic data only instead of real target-site data (ablation)Polar Transformer · Same-site solar elevationmedian RMSEImproved
use physics-guided synthetic training sequences with Polar Transformer instead of prior baselinesPolar Transformer · Cross-site, site-averagemedian geodesic error (km): 513 km vs 2,394 km (SecTran-MIM) and 3,971 km (RI-ResNet-RDM)Improved

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