AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

Orthogonal Spatial-Aware Multi-View Anchor Graph Clustering for Incomplete Remote Sensing Data

Yongshan Zhang, Xiaohuan Lin, Lefei Zhang, Zhihua Cai

CVPR 2026 · Graph learning › Graph algorithms and clustering · approach: Anchor-based multi-view clustering

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 superpixel anchoring and graph initialization (SAGI) instead of k-means anchors (ablation)OSMAGC · MDAS (0.5 missing)ACC: 0.3220 -> 0.4345Improved
use multi-scale anchor graph learning instead of one superpixel size per view (ablation)OSMAGC · Augsburg (0.5 missing)ACC: 0.5530 -> 0.6120Improved
learn alignment matrices (SACFL) instead of fixing them (ablation)OSMAGC · MUUFL (0.5 missing)ACC: 0.4463 -> 0.5002Improved
add orthogonal spatial-aware regularization (OSAR) (ablation)OSMAGC · Berlin (0.5 missing)ACC: 0.4375 -> 0.4410Improved
add OSAR (ablation)OSMAGC · Augsburg (0.5 missing)ACC: 0.5689 -> 0.6120Improved
use OSMAGC instead of best baselineOSMAGC · MUUFL (0.5 missing)ACC: 0.4036 -> 0.5002Improved
use OSMAGC instead of best baselineOSMAGC · MDAS (0.5 missing)ACC: 0.4171 -> 0.4345Improved
use more superpixels (ablation)OSMAGC · Berlinclustering performanceHurt
use more superpixels (ablation)OSMAGC · MUUFLclustering performanceImproved
use large lambda (ablation)OSMAGC · Berlinclustering performanceHurt

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