AI Paper Atlas

Research map / 3D vision / Depth and stereo

LinStereo: Linear-Complexity Global Attention for Multi-Scale Iterative Stereo Matching

Yiran Wang, Oliver Turner, Viorela Ila

ECCV 2026 · 3D vision › Depth and stereo · approach: Learned stereo with specialized sensors · 2 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
replace ConvGRU with global linear attention plus positional encodings and gating (full PALA vs plain linear attention) (ablation)LinStereo · KITTI 2015EPE: 1.18 -> 1.01Improved
use asymmetric rather than symmetric RoPE (ablation)PALA · KITTI 2015EPE: 1.05 -> 1.01Improved
add PALA to HSCV+DPI (ablation)LinStereo · KITTI 2015EPE: 1.21 -> 1.01Improved
use per-scale hierarchical cost volumes instead of pooled multi-scale feature (ablation)LinStereo · KITTI 2015EPE: 1.07 -> 1.01Improved
add Depth Prior Initialization (ablation)LinStereo · KITTI 2015EPE: 1.10 -> 1.01Improved
raise iterations from 8 to 12 (ablation)LinStereo · KITTI 2015EPE: 1.01 vs 1.03No clear effect
reduce iterations from 8 to 2 (ablation)LinStereo · KITTI 2015EPE: 1.01 -> 1.24Hurt
use 3 update scales instead of 1 (ablation)LinStereo · KITTI 2015EPE: 1.13 -> 1.01Improved
stack more PALA blocks per updater (ablation)PALA · KITTI 2015EPE: 1.01 -> 1.05Hurt
learning rate 4e-4 instead of 2e-4 (ablation)LinStereo · KITTI 2015EPE: 1.01 -> 1.16Hurt
PALA vs ConvGRU update operator latencyLinStereo · per-iteration latencyms: 3.50 vs 3.63 and 3.43No clear effect
global linear-attention decoder on VFM backbone (SceneFlow only) vs larger extra-data methodsLinStereo · TartanAir-UWAbsRel: 0.05 -> 0.04Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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