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 2204.13630cs.CV
收藏

旋转等变三维目标检测

Rotationally Equivariant 3D Object Detection

Stanford University(斯坦福大学);Tsinghua University(清华大学);Shanghai Qi Zhi Institute(上海期智研究院)

Hong-Xing Yu, Jiajun Wu, Li Yi

摘要:

旋转等方差最近已经成为3D深度学习社区中强烈期望的属性。然而,大多数现有的方法都集中在关于全局输入旋转的等方差上,而忽略了旋转对称性具有其自身的空间支持的事实。具体来说,我们考虑的目标检测问题,在3D场景中,一个对象的限界框应该是等变的对象姿态,独立的场景运动。这表明了一个新的期望属性,我们称之为对象级旋转等方差。为了将对象级旋转等变合并到3D对象检测器中,我们需要一种机制来提取具有局部对象级空间支持的等变特征,同时能够对跨对象上下文信息进行建模。为此,我们提出了等变目标检测网络(EON)与旋转等方差悬浮设计,以实现对象级的等方差。EON可以应用于现代点云对象检测器,如VoteNet和PointRCNN,使它们能够在场景尺度输入中利用对象旋转对称性。我们在室内场景和自动驾驶数据集上的实验表明,通过将我们的EON设计插入到现有的最先进的3D物体检测器中,可以获得显着的改进。

英文摘要:

Rotation equivariance has recently become a strongly desired property in the 3D deep learning community. Yet most existing methods focus on equivariance regarding a global input rotation while ignoring the fact that rotation symmetry has its own spatial support. Specifically, we consider the object detection problem in 3D scenes, where an object bounding box should be equivariant regarding the object pose, independent of the scene motion. This suggests a new desired property we call object-level rotation equivariance. To incorporate object-level rotation equivariance into 3D object detectors, we need a mechanism to extract equivariant features with local object-level spatial support while being able to model cross-object context information. To this end, we propose Equivariant Object detection Network (EON) with a rotation equivariance suspension design to achieve object-level equivariance. EON can be applied to modern point cloud object detectors, such as VoteNet and PointRCNN, enabling them to exploit object rotation symmetry in scene-scale inputs. Our experiments on both indoor scene and autonomous driving datasets show that significant improvements are obtained by plugging our EON design into existing state-of-the-art 3D object detectors.

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。

↑