AI Paper Atlas

Research map / Detection and segmentation / CNN features

WovenAnchor Matcher: Specialized Intra- and Inter-Image Context Modeling for Feature Matching

Zhiyang Li, Ruijiang Jin, Thibaut Klenke, Yusuke Sekikawa, Nakamasa Inoue

NeurIPS 2026 · Detection and segmentation › CNN features · approach: Local features and context in CNNs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes WovenAnchor Matcher, a semi-dense feature matcher that separates intra-image context modeling (preserving 2D structure) from inter-image aggregation (gathering from plausible corresponding regions while limiting noise).

Problem

Semi-dense local feature matching aggregates context before coarse-to-fine correspondence estimation, but intra-image and inter-image aggregation serve different roles and are commonly treated uniformly.

Approach

Specialize the context modeling per role: intra-image aggregation propagates spatial evidence while retaining structured 2D dependencies, and inter-image aggregation gathers information only from plausible corresponding regions to limit noise.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add WovenMamba intra-image aggregation (ablation)WAM · MegaDepth-1500pose AUC@5deg: 64.6 -> 66.1Improved
add AnchorCrossAttention inter-image aggregation (ablation)WAM · MegaDepth-1500pose AUC@5deg: 64.0 -> 66.1Improved
use both specialized components instead of neither (ablation)WAM · MegaDepth-1500pose AUC@5deg: 62.0 -> 66.1Improved
use horizontal-then-vertical WovenMamba instead of independent directional scans (FourMamba) (ablation)WAM intra-image module · MegaDepth-1500pose AUC@5deg: 65.3 -> 66.1Improved
use WovenMamba instead of self-attention for intra-image (ablation)WAM · MegaDepth-1500pose AUC@5deg: 64.8 -> 66.1Improved
use anchor cross-attention instead of dense cross-attention (ablation)WAM inter-image module · MegaDepth-1500pose AUC@5deg: 65.0 -> 66.1Improved
assign WovenMamba to intra and ACA to inter roles instead of reversed or shared roles (ablation)WAM · MegaDepth-1500pose AUC@5deg: 64.9 -> 66.1Improved
use learned instead of uniform anchor assignments (ablation)ACA · MegaDepth-1500pose AUC@5deg: 65.7 -> 66.1Improved
add anchor geometry features (ablation)ACA · MegaDepth-1500pose AUC@5deg: 65.8 -> 66.1Improved
attend jointly across all anchor IDs (ablation)ACA · MegaDepth-1500pose AUC@5deg: 66.1 -> 66.3Improved
increase anchors per window beyond 4 (ablation)ACA · MegaDepth-1500pose AUC@5deg: 66.1 -> 65.7No clear effect
add WovenMamba separator tokens (ablation)WAM · ScanNet-1500 zero-shotpose AUC@5deg: 22.5 -> 23.2Improved

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