AI Paper Atlas

Research map / Detection and segmentation / 3D detection and driving

Seeing in Double: Dual-Granularity BEV Segmentation via Mamba-Driven Alignment and Polar-Decoupled Experts

Jiaxin Cai, Rui Lin, Jingze Su, Qi Li, Wenjie Yang, Yuanlong Yu, Wenxi Liu

AAAI 2026 · Detection and segmentation › 3D detection and driving · approach: Multi-camera BEV detection · 1 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 dual-resolution heterogeneous network (ablation)BEV baseline · nuScenes ablation (visibility filtering, 224x448)IoU: 42.0 -> 43.9Improved
add MRADE height estimation (ablation)BEV baseline · nuScenes ablation (visibility filtering, 224x448)IoU: 42.0 -> 43.1Improved
add MCRA alignment (ablation)full model · nuScenes ablation (visibility filtering, 224x448)IoU: 45.3 -> 46.1Improved
add ABSS without MCRA (ablation)dual-res + MRADE · nuScenes ablation (visibility filtering, 224x448)IoU: 44.5 -> 45.3Improved
shared-state Mamba alignment instead of cross-attention (ablation)MCRA · nuScenes ablation (visibility filtering, 224x448)IoU: 45.5 -> 46.1Improved
shared instead of independent Mamba state spaces (ablation)MCRA · nuScenes ablation (visibility filtering, 224x448)IoU: 45.8 -> 46.1Improved
add permanent expert to MRADE (ablation)MRADE · nuScenes ablation (visibility filtering, 224x448)IoU: 45.5 -> 46.1Improved
decouple radial and angular experts (ablation)MRADE · nuScenes ablation (visibility filtering, 224x448)IoU: 45.6 -> 46.1Improved
MRADE instead of fixed height bins (ablation)height estimation · nuScenes ablation (visibility filtering, 224x448)IoU: 44.7 -> 46.1Improved
full method vs PointBeVOurs · nuScenes vehicle, no visibility filtering, 224x480IoU: 38.7 -> 42.4Improved
full method vs PointBeV on Lyft L5 short rangeOurs · Lyft L5 short rangeIoU: 72.6 -> 73.3Improved
full method vs PointBeV runtimeOurs · nuScenes efficiencyFPS: 15 -> 14 FPS; memory 1.21 -> 1.58 GMixed

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