PF-Net: Point Fractal Network for 3D Point Cloud Completion

By Zitian Huang, Yikuan Yu, Jiawen Xu, Feng Ni, Xinyi Le. 2020

475 citations recorded by the source corpus.

In this paper, we propose a Point Fractal Network (PF-Net), a novel learning-based approach for precise and high-fidelity point cloud completion. Unlike existing point cloud completion networks, which generate the overall shape of the point cloud from the incomplete point cloud and always change existing points and encounter noise and geometrical loss, PF-Net preserves the spatial arrangements of the incomplete point cloud and can figure out the detailed geometrical structure of the missing region(s) in the prediction. To succeed at this task, PF-Net estimates the missing point cloud hierarchically by utilizing a feature-points-based multi-scale generating network. Further, we add up multi-stage completion loss and adversarial loss to generate more realistic missing region(s). The adversarial loss can better tackle multiple modes in the prediction. Our experiments demonstrate the effectiveness of our method for several challenging point cloud completion tasks.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, 3D Vision & Pose

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

In 3D Vision & Pose

Citations in the corpus

Multiview Compressive Coding for 3D Reconstruction (2023, methodology): “For objects, methods directly output full point clouds [21,79,80] or deploy generative models [76, 84], but are typically tied to a fixed resolution.”
SnowflakeNet: Point Cloud Completion by Snowflake Point Deconvolution with Skip-Transformer (2021, background): “Followers like CDN [46] and PF-Net [27] further extended the number of generation stages and achieved the currently state-of-theart performance.”
PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers (2021, methodology): “Following previous work [16], we only predict the missing parts of the point cloud and concatenate them with the input point cloud to obtain the complete objects.”
Revealing Occlusions with 4D Neural Fields (2022, background): “Our goal is somewhat similar to that of point completion networks [21, 55, 60, 62, 68], although these works typically operate on a per-object basis and address only self-occlusions or amodal completion.”
Unsupervised 3D Shape Completion through GAN Inversion (2021, methodology): “The ball-holed partial shapes in this section are generated as described in PF-Net [15], i.e., removing exactly 512 points from the complete shape.”