LESV: Language Embedded Sparse Voxel Fusion for Open-Vocabulary 3D Scene Understanding
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914439195787264 |
|---|---|
| author | Wang, Fusang Piasco, Nathan Bennehar, Moussab Roldão, Luis Tsishkou, Dzmitry Moutarde, Fabien |
| author_facet | Wang, Fusang Piasco, Nathan Bennehar, Moussab Roldão, Luis Tsishkou, Dzmitry Moutarde, Fabien |
| contents | Recent advancements in open-vocabulary 3D scene understanding heavily rely on 3D Gaussian Splatting (3DGS) to register vision-language features into 3D space. However, we identify two critical limitations in these approaches: the spatial ambiguity arising from unstructured, overlapping Gaussians which necessitates probabilistic feature registration, and the multi-level semantic ambiguity caused by pooling features over object-level masks, which dilutes fine-grained details. To address these challenges, we present a novel framework that leverages Sparse Voxel Rasterization (SVRaster) as a structured, disjoint geometry representation. By regularizing SVRaster with monocular depth and normal priors, we establish a stable geometric foundation. This enables a deterministic, confidence-aware feature registration process and suppresses the semantic bleeding artifact common in 3DGS. Furthermore, we resolve multi-level ambiguity by exploiting the emerging dense alignment properties of foundation model AM-RADIO, avoiding the computational overhead of hierarchical training methods. Our approach achieves state-of-the-art performance on Open Vocabulary 3D Object Retrieval and Point Cloud Understanding benchmarks, particularly excelling on fine-grained queries where registration methods typically fail. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_01388 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LESV: Language Embedded Sparse Voxel Fusion for Open-Vocabulary 3D Scene Understanding Wang, Fusang Piasco, Nathan Bennehar, Moussab Roldão, Luis Tsishkou, Dzmitry Moutarde, Fabien Computer Vision and Pattern Recognition I.4.5, I.2.10 Recent advancements in open-vocabulary 3D scene understanding heavily rely on 3D Gaussian Splatting (3DGS) to register vision-language features into 3D space. However, we identify two critical limitations in these approaches: the spatial ambiguity arising from unstructured, overlapping Gaussians which necessitates probabilistic feature registration, and the multi-level semantic ambiguity caused by pooling features over object-level masks, which dilutes fine-grained details. To address these challenges, we present a novel framework that leverages Sparse Voxel Rasterization (SVRaster) as a structured, disjoint geometry representation. By regularizing SVRaster with monocular depth and normal priors, we establish a stable geometric foundation. This enables a deterministic, confidence-aware feature registration process and suppresses the semantic bleeding artifact common in 3DGS. Furthermore, we resolve multi-level ambiguity by exploiting the emerging dense alignment properties of foundation model AM-RADIO, avoiding the computational overhead of hierarchical training methods. Our approach achieves state-of-the-art performance on Open Vocabulary 3D Object Retrieval and Point Cloud Understanding benchmarks, particularly excelling on fine-grained queries where registration methods typically fail. |
| title | LESV: Language Embedded Sparse Voxel Fusion for Open-Vocabulary 3D Scene Understanding |
| topic | Computer Vision and Pattern Recognition I.4.5, I.2.10 |
| url | https://arxiv.org/abs/2604.01388 |