dorsal/arxiv
View SchemaVisual Saliency and Attention as Random Walks on Complex Networks
| Authors | Luciano da Fontoura Costa |
|---|---|
| Categories | |
| ArXiv ID | physics/0603025 |
| URL | https://arxiv.org/abs/physics/0603025 |
Abstract
The current article shows how concepts from the areas of random walks, Markov chains, complex networks and image analysis can be naturally combined in order to provide a unified and biologically plausible model relating saliency and visual attention. Two types of models are proposed: (i) images are converted into complex networks by considering pixels as nodes while connections are established in terms of fields of influence defined by visual features such as tangent fields induced by gray-level contrasts and distance; and (ii) image pixels exhibiting particularly distinctive values of visual properties such as gray-level intensity, contrast, size of objects, orientation and texture are mapped into nodes and the weights of links are defined in order to favor transitions between regions with similar or different visual features, also taking the distance between the nodes into account. Preferential random walks are performed on such networks in order to emulate attentional shifts and eye movements, and the saliency of each region is obtained in terms of the frequency of visits to each node at equilibrium. In the case of the first model, there is a definite tendency to emphasize not only high curvature points but also convergences of the tangent field. The frequency of visits is found to be strongly correlated with the node degrees (strengths) for this model. Different results have been obtained for the second model as a consequence of the directed and asymmetric nature of the respectively obtained networks.
{
"annotation_id": "ebff0c02-7094-4768-b2ed-0f1a222a50a8",
"date_created": "2026-03-02T18:01:06.620000Z",
"date_modified": "2026-03-02T18:01:06.620000Z",
"file_hash": "01dc3d972f2d43762419cc86faf90e707b8e28940e21f67b4f6bae9436bbf300",
"private": false,
"record": {
"abstract": "The current article shows how concepts from the areas of random walks, Markov\nchains, complex networks and image analysis can be naturally combined in order\nto provide a unified and biologically plausible model relating saliency and\nvisual attention. Two types of models are proposed: (i) images are converted\ninto complex networks by considering pixels as nodes while connections are\nestablished in terms of fields of influence defined by visual features such as\ntangent fields induced by gray-level contrasts and distance; and (ii) image\npixels exhibiting particularly distinctive values of visual properties such as\ngray-level intensity, contrast, size of objects, orientation and texture are\nmapped into nodes and the weights of links are defined in order to favor\ntransitions between regions with similar or different visual features, also\ntaking the distance between the nodes into account. Preferential random walks\nare performed on such networks in order to emulate attentional shifts and eye\nmovements, and the saliency of each region is obtained in terms of the\nfrequency of visits to each node at equilibrium. In the case of the first\nmodel, there is a definite tendency to emphasize not only high curvature points\nbut also convergences of the tangent field. The frequency of visits is found to\nbe strongly correlated with the node degrees (strengths) for this model.\nDifferent results have been obtained for the second model as a consequence of\nthe directed and asymmetric nature of the respectively obtained networks.",
"arxiv_id": "physics/0603025",
"authors": [
"Luciano da Fontoura Costa"
],
"categories": [
"physics.med-ph",
"cond-mat.dis-nn"
],
"title": "Visual Saliency and Attention as Random Walks on Complex Networks",
"url": "https://arxiv.org/abs/physics/0603025"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "3d59a20f-bd82-4c00-bda7-88cadeb8fd4f",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}