arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.02508cs.CV
收藏

使用Vote2Cap-DETR实现端到端3D密集字幕

End-to-End 3D Dense Captioning with Vote2Cap-DETR

Fudan University(复旦大学);Tencent PCG(腾讯平台与内容事业群);Sichuan University(四川大学)

Sijin Chen, Hongyuan Zhu, Xin Chen, Yinjie Lei, Tao Chen, Gang YU

摘要:

3D密集字幕旨在生成与其相关联的对象区域定位的多个字幕。现有的方法遵循一个复杂的“检测然后描述”管道,配备了许多手工制作的组件。然而,这些手工制作的组件将产生次优的性能,给定杂乱的对象空间和不同场景之间的类分布。在本文中,我们提出了一个简单而有效的Transformer框架Vote 2Cap-DETR的基础上最近流行的\textbf{DE}保护\textbf{TR} ansformer(DETR)。与现有技术相比,我们的框架具有几个吸引人的优点:1)不诉诸于大量的手工制作的组件,我们的方法是基于一个完整的Transformer编码器-解码器架构,具有可学习的投票查询驱动的对象解码器,以及以集合预测方式产生密集字幕的字幕解码器。2)与两阶段方案相比,我们的方法可以在一个阶段中执行检测和字幕。3)在没有花里胡哨的情况下,在两个常用数据集ScanRefer和Nr 3D上进行的广泛实验表明,我们的Vote 2Cap-DETR在CIDEr@0.5IoU中分别超过了当前最先进的11.13%和7.11%。代码将很快发布。

英文摘要:

3D dense captioning aims to generate multiple captions localized with their associated object regions. Existing methods follow a sophisticated ``detect-then-describe'' pipeline equipped with numerous hand-crafted components. However, these hand-crafted components would yield suboptimal performance given cluttered object spatial and class distributions among different scenes. In this paper, we propose a simple-yet-effective transformer framework Vote2Cap-DETR based on recent popular \textbf{DE}tection \textbf{TR}ansformer (DETR). Compared with prior arts, our framework has several appealing advantages: 1) Without resorting to numerous hand-crafted components, our method is based on a full transformer encoder-decoder architecture with a learnable vote query driven object decoder, and a caption decoder that produces the dense captions in a set-prediction manner. 2) In contrast to the two-stage scheme, our method can perform detection and captioning in one-stage. 3) Without bells and whistles, extensive experiments on two commonly used datasets, ScanRefer and Nr3D, demonstrate that our Vote2Cap-DETR surpasses current state-of-the-arts by 11.13\% and 7.11\% in CIDEr@0.5IoU, respectively. Codes will be released soon.

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。

↑