dorsal/arxiv
View SchemaObjSplat: Geometry-Aware Gaussian Surfels for Active Object Reconstruction
| Authors | Yuetao Li, Zhizhou Jia, Yu Zhang, Qun Hao, Shaohui Zhang |
|---|---|
| Categories | |
| ArXiv ID | 2601.06997vv1 |
| URL | https://arxiv.org/abs/2601.06997 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Autonomous high-fidelity object reconstruction is fundamental for creating digital assets and bridging the simulation-to-reality gap in robotics. We present ObjSplat, an active reconstruction framework that leverages Gaussian surfels as a unified representation to progressively reconstruct unknown objects with both photorealistic appearance and accurate geometry. Addressing the limitations of conventional opacity or depth-based cues, we introduce a geometry-aware viewpoint evaluation pipeline that explicitly models back-face visibility and occlusion-aware multi-view covisibility, reliably identifying under-reconstructed regions even on geometrically complex objects. Furthermore, to overcome the limitations of greedy planning strategies, ObjSplat employs a next-best-path (NBP) planner that performs multi-step lookahead on a dynamically constructed spatial graph. By jointly optimizing information gain and movement cost, this planner generates globally efficient trajectories. Extensive experiments in simulation and on real-world cultural artifacts demonstrate that ObjSplat produces physically consistent models within minutes, achieving superior reconstruction fidelity and surface completeness while significantly reducing scan time and path length compared to state-of-the-art approaches. Project page: https://li-yuetao.github.io/ObjSplat-page/ .
{
"annotation_id": "724d016f-0b19-40db-afbd-ba5f2c5161fa",
"date_created": "2026-02-17T05:53:08.755000Z",
"date_modified": "2026-02-17T05:53:08.755000Z",
"file_hash": "078ed0b8a081ee2c24c5e756f82e334d47de8d13e83e77b80fc5487e6edc625f",
"private": false,
"record": {
"abstract": "Autonomous high-fidelity object reconstruction is fundamental for creating digital assets and bridging the simulation-to-reality gap in robotics. We present ObjSplat, an active reconstruction framework that leverages Gaussian surfels as a unified representation to progressively reconstruct unknown objects with both photorealistic appearance and accurate geometry. Addressing the limitations of conventional opacity or depth-based cues, we introduce a geometry-aware viewpoint evaluation pipeline that explicitly models back-face visibility and occlusion-aware multi-view covisibility, reliably identifying under-reconstructed regions even on geometrically complex objects. Furthermore, to overcome the limitations of greedy planning strategies, ObjSplat employs a next-best-path (NBP) planner that performs multi-step lookahead on a dynamically constructed spatial graph. By jointly optimizing information gain and movement cost, this planner generates globally efficient trajectories. Extensive experiments in simulation and on real-world cultural artifacts demonstrate that ObjSplat produces physically consistent models within minutes, achieving superior reconstruction fidelity and surface completeness while significantly reducing scan time and path length compared to state-of-the-art approaches. Project page: https://li-yuetao.github.io/ObjSplat-page/ .",
"arxiv_id": "2601.06997",
"authors": [
"Yuetao Li",
"Zhizhou Jia",
"Yu Zhang",
"Qun Hao",
"Shaohui Zhang"
],
"categories": [
"cs.RO",
"cs.CV"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "ObjSplat: Geometry-Aware Gaussian Surfels for Active Object Reconstruction",
"url": "https://arxiv.org/abs/2601.06997",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "f3a8c9da-5f74-4211-a9c3-42c16fb29e56",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}