arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.00544cs.CVcs.AI
收藏

面向视觉任务的深度外在流形表示

Deep Extrinsic Manifold Representation for Vision Tasks

ecnu(华东师范大学)

Tongtong Zhang, Xian Wei, Yuanxiang Li

AI总结:

本文提出深度外在流形表示 DEMR,通过外在嵌入避免直接优化测地线损失,在 SE(3) 点云对齐和 Grassmann 流形光照子空间学习上验证了其可行性与泛化能力。

摘要:

非欧几里得数据在不同领域中频繁出现,但针对以流形表示为输出训练神经网络这一根本挑战的文献仍然有限。在此背景下,我们为视觉任务引入了一种名为深度外在流形表示(DEMR)的方法。DEMR 将外在流形嵌入引入深度神经网络,从而帮助生成流形表示。DEMR 方法并不直接优化复杂的测地线损失,而是专注于优化嵌入欧几里得空间中的计算图,使其能够适应各种架构需求。我们在两类流形上提供了支持该概念的经验证据,即 SE(3) 及其相关的商流形。这些证据在可行性、渐近性质和泛化能力方面提供了理论保证。实验结果表明,DEMR 能够有效适应点云对齐,在 SE(3) 中产生输出,也适用于输出位于 Grassmann 流形上的光照子空间学习。

英文摘要:

Non-Euclidean data is frequently encountered across different fields, yet there is limited literature that addresses the fundamental challenge of training neural networks with manifold representations as outputs. We introduce the trick named Deep Extrinsic Manifold Representation (DEMR) for visual tasks in this context. DEMR incorporates extrinsic manifold embedding into deep neural networks, which helps generate manifold representations. The DEMR approach does not directly optimize the complex geodesic loss. Instead, it focuses on optimizing the computation graph within the embedded Euclidean space, allowing for adaptability to various architectural requirements. We provide empirical evidence supporting the proposed concept on two types of manifolds, $SE(3)$ and its associated quotient manifolds. This evidence offers theoretical assurances regarding feasibility, asymptotic properties, and generalization capability. The experimental results show that DEMR effectively adapts to point cloud alignment, producing outputs in $ SE(3) $, as well as in illumination subspace learning with outputs on the Grassmann manifold.

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。

↑