Problem
Research questions and friction points this paper is trying to address.
adjacency matrices
node-link images
graph information
visual observations
dependencies among edges
VisAdj: Learning Adjacency Matrices from Node-Link Images
๐ค AI Summary
ๆฌๆๆๅบVisAdjๆกๆถ๏ผ้่ฟ่ช้ๅบ้ๆฉๅ้่็นๅฏนๅ่ๅ่พน็ผๆจ็่งฃๅณไป่็น้พๆฅๅพๅไธญๅญฆไน ้ปๆฅ็ฉ้ต็้ฎ้ขใ
๐ Abstract
Learning adjacency matrices from node-link images is a fundamental problem for recovering structured graph information from visual observations. Existing methods typically rely on fixed KNN-based heuristics for candidate edge selection and fail to capture dependencies among edges. To overcome these limitations, we propose VisAdj, a new framework for topology-aware adjacency prediction. VisAdj introduces an attention-sparse neighbor sampler to adaptively select a high-recall set of candidate node pairs and performs joint edge inference using a line-graph transformer that treats candidate edges as tokens and explicitly models dependencies among incident edges. Extensive experiments on synthetic graphs, road networks, and vessel images demonstrate that VisAdj consistently outperforms existing baselines by clear margins.
Innovation
Methods, ideas, or system contributions that make the work stand out.
attention-sparse neighbor sampler
line-graph transformer
topology-aware adjacency prediction
๐ Similar Papers
No similar papers found.
