SnowflakeNet: Point Cloud Completion by Snowflake Point Deconvolution with Skip-Transformer
By Peng Xiang, Xin Wen, Yu-Shen Liu, Yan-Pei Cao, Pengfei Wan, Wen Zheng, Zhizhong Han. IEEE International Conference on Computer Vision, 2021
340 citations recorded by the source corpus.
Point cloud completion aims to predict a complete shape in high accuracy from its partial observation. However, previous methods usually suffered from discrete nature of point cloud and unstructured prediction of points in local regions, which makes it hard to reveal fine local geometric details on the complete shape. To resolve this issue, we propose SnowflakeNet with Snowflake Point Deconvolution (SPD) to generate the complete point clouds. The SnowflakeNet models the generation of complete point clouds as the snowflake-like growth of points in 3D space, where the child points are progressively generated by splitting their parent points after each SPD. Our insight of revealing detailed geometry is to introduce skip-transformer in SPD to learn point splitting patterns which can fit local regions the best. Skip-transformer leverages attention mechanism to summarize the splitting patterns used in the previous SPD layer to produce the splitting in the current SPD layer. The locally compact and structured point cloud generated by SPD is able to precisely capture the structure characteristic of 3D shape in local patches, which enables the network to predict highly detailed geometries, such as smooth regions, sharp edges and corners. Our experimental results outperform the state-of-the-art point cloud completion methods under widely used benchmarks. Code will be available at https://github.com/AllenXiangX/SnowflakeNet.
Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/dbc9e73b75c205d493e0a7242b038e5c6ac04efe
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
Citations in the corpus
- Triplane Meets Gaussian Splatting: Fast and Generalizable Single-View 3D Reconstruction with Transformers (2023, positive): “Therefore, we adopt a lifting module with two-step Snowflake point deconvolution (SPD) [73, 74] to densify the point clouds from 2048 points to 16384 points.”
- A Survey of Visual Transformers (2021, background): “SnowflakeNet [124] formulates the process of completing point clouds as a snowflake-like growth, which progressively generates child points from their parent points implemented by a pointwise splitting deconvolution strategy.”