dorsal/arxiv
View SchemaSAM3-DMS: Decoupled Memory Selection for Multi-target Video Segmentation of SAM3
| Authors | Ruiqi Shen, Chang Liu, Henghui Ding |
|---|---|
| Categories | |
| ArXiv ID | 2601.09699vv1 |
| URL | https://arxiv.org/abs/2601.09699 |
| License | http://creativecommons.org/licenses/by-nc-sa/4.0/ |
Abstract
Segment Anything 3 (SAM3) has established a powerful foundation that robustly detects, segments, and tracks specified targets in videos. However, in its original implementation, its group-level collective memory selection is suboptimal for complex multi-object scenarios, as it employs a synchronized decision across all concurrent targets conditioned on their average performance, often overlooking individual reliability. To this end, we propose SAM3-DMS, a training-free decoupled strategy that utilizes fine-grained memory selection on individual objects. Experiments demonstrate that our approach achieves robust identity preservation and tracking stability. Notably, our advantage becomes more pronounced with increased target density, establishing a solid foundation for simultaneous multi-target video segmentation in the wild.
{
"annotation_id": "982b265b-c563-4236-83c0-71bc55b2dc0c",
"date_created": "2026-02-17T05:53:20.178000Z",
"date_modified": "2026-02-17T05:53:20.178000Z",
"file_hash": "708bf71dac509dd60d4beed927008c07bb28e3d4c27d3dad3381a4c4a99cb8b1",
"private": false,
"record": {
"abstract": "Segment Anything 3 (SAM3) has established a powerful foundation that robustly detects, segments, and tracks specified targets in videos. However, in its original implementation, its group-level collective memory selection is suboptimal for complex multi-object scenarios, as it employs a synchronized decision across all concurrent targets conditioned on their average performance, often overlooking individual reliability. To this end, we propose SAM3-DMS, a training-free decoupled strategy that utilizes fine-grained memory selection on individual objects. Experiments demonstrate that our approach achieves robust identity preservation and tracking stability. Notably, our advantage becomes more pronounced with increased target density, establishing a solid foundation for simultaneous multi-target video segmentation in the wild.",
"arxiv_id": "2601.09699",
"authors": [
"Ruiqi Shen",
"Chang Liu",
"Henghui Ding"
],
"categories": [
"cs.CV"
],
"license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
"title": "SAM3-DMS: Decoupled Memory Selection for Multi-target Video Segmentation of SAM3",
"url": "https://arxiv.org/abs/2601.09699",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "92fb44f8-c473-46e9-b60c-ee64c4ac0479",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}