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 2007.10170cs.CV
收藏

用于高效点云生成的离散点流网络

Discrete Point Flow Networks for Efficient Point Cloud Generation

Univ. Grenoble Alpes(格勒诺布尔-阿尔卑斯大学);Inria(法国国家信息与自动化研究所);CNRS(法国国家科学研究中心);Grenoble INP(格勒诺布尔理工学院);Facebook AI Research(Facebook人工智能研究院)

Roman Klokov, Edmond Boyer, Jakob Verbeek

摘要:

生成模型已被证明在建模3D形状及其统计变化方面是有效的。在本文中,我们调查他们的应用点云,三维形状表示广泛用于计算机视觉,但是,只有少数生成模型尚未提出。我们引入了一个潜在的变量模型,建立在规范化流与仿射耦合层生成一个潜在的形状表示的任意大小的3D点云。为了评估其形状建模的好处,我们将此模型用于生成,自动编码和单视图形状重建任务。我们在评估生成和自动编码的大多数指标方面改进了最近的基于GAN的模型。与最近基于连续流的工作相比,我们的模型在训练和推理时间上都有显着的加速,以获得类似或更好的性能。对于单视图形状重建,我们还获得了与最先进的体素,点云和基于网格的方法相当的结果。

英文摘要:

Generative models have proven effective at modeling 3D shapes and their statistical variations. In this paper we investigate their application to point clouds, a 3D shape representation widely used in computer vision for which, however, only few generative models have yet been proposed. We introduce a latent variable model that builds on normalizing flows with affine coupling layers to generate 3D point clouds of an arbitrary size given a latent shape representation. To evaluate its benefits for shape modeling we apply this model for generation, autoencoding, and single-view shape reconstruction tasks. We improve over recent GAN-based models in terms of most metrics that assess generation and autoencoding. Compared to recent work based on continuous flows, our model offers a significant speedup in both training and inference times for similar or better performance. For single-view shape reconstruction we also obtain results on par with state-of-the-art voxel, point cloud, and mesh-based methods.

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。

↑