arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.08907cs.CVcs.MM
收藏

用于3D视觉定位的双属性-空间关系对齐

Dual Attribute-Spatial Relation Alignment for 3D Visual Grounding

School of Information Science;Technology University of Science;School of Data Science University of Science;Department of Computer Science University of Rochester New York, United States

Yue Xu, Kaizhi Yang, Jiebo Luo, Xuejin Chen

AI总结:

针对3D视觉定位,本文提出DASANet,通过双分支注意力与交叉注意力分别对齐语言和3D点云中的属性及空间关系,在Nr3D上达到65.1%的最佳准确率。

摘要:

3D视觉定位是一个新兴研究领域,致力于建立3D物理世界与自然语言之间的联系,这对于实现具身智能至关重要。本文提出DASANet,这是一个双属性-空间关系对齐网络,分别建模并对齐语言与3D视觉模态之间的对象属性和空间关系特征。我们将语言和3D点云输入都分解为两个独立部分,并设计双分支注意力模块来分别建模分解后的输入,同时通过交叉注意力在属性-空间特征融合中保留全局上下文。我们的DASANet在Nr3D数据集上取得65.1%的最高定位准确率,比最佳竞争方法高出1.3%。此外,两个分支的可视化证明了该方法的高效性和高度可解释性。

英文摘要:

3D visual grounding is an emerging research area dedicated to making connections between the 3D physical world and natural language, which is crucial for achieving embodied intelligence. In this paper, we propose DASANet, a Dual Attribute-Spatial relation Alignment Network that separately models and aligns object attributes and spatial relation features between language and 3D vision modalities. We decompose both the language and 3D point cloud input into two separate parts and design a dual-branch attention module to separately model the decomposed inputs while preserving global context in attribute-spatial feature fusion by cross attentions. Our DASANet achieves the highest grounding accuracy 65.1% on the Nr3D dataset, 1.3% higher than the best competitor. Besides, the visualization of the two branches proves that our method is efficient and highly interpretable.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑