Saved in:
Bibliographic Details
Main Authors: Zhang, Tianyu, Qian, Guocheng, Xie, Jin, Yang, Jian
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.19573
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910667386126336
author Zhang, Tianyu
Qian, Guocheng
Xie, Jin
Yang, Jian
author_facet Zhang, Tianyu
Qian, Guocheng
Xie, Jin
Yang, Jian
contents Point cloud frame interpolation is a challenging task that involves accurate scene flow estimation across frames and maintaining the geometry structure. Prevailing techniques often rely on pre-trained motion estimators or intensive testing-time optimization, resulting in compromised interpolation accuracy or prolonged inference. This work presents FastPCI that introduces Pyramid Convolution-Transformer architecture for point cloud frame interpolation. Our hybrid Convolution-Transformer improves the local and long-range feature learning, while the pyramid network offers multilevel features and reduces the computation. In addition, FastPCI proposes a unique Dual-Direction Motion-Structure block for more accurate scene flow estimation. Our design is motivated by two facts: (1) accurate scene flow preserves 3D structure, and (2) point cloud at the previous timestep should be reconstructable using reverse motion from future timestep. Extensive experiments show that FastPCI significantly outperforms the state-of-the-art PointINet and NeuralPCI with notable gains (e.g. 26.6% and 18.3% reduction in Chamfer Distance in KITTI), while being more than 10x and 600x faster, respectively. Code is available at https://github.com/genuszty/FastPCI
format Preprint
id arxiv_https___arxiv_org_abs_2410_19573
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FastPCI: Motion-Structure Guided Fast Point Cloud Frame Interpolation
Zhang, Tianyu
Qian, Guocheng
Xie, Jin
Yang, Jian
Computer Vision and Pattern Recognition
Point cloud frame interpolation is a challenging task that involves accurate scene flow estimation across frames and maintaining the geometry structure. Prevailing techniques often rely on pre-trained motion estimators or intensive testing-time optimization, resulting in compromised interpolation accuracy or prolonged inference. This work presents FastPCI that introduces Pyramid Convolution-Transformer architecture for point cloud frame interpolation. Our hybrid Convolution-Transformer improves the local and long-range feature learning, while the pyramid network offers multilevel features and reduces the computation. In addition, FastPCI proposes a unique Dual-Direction Motion-Structure block for more accurate scene flow estimation. Our design is motivated by two facts: (1) accurate scene flow preserves 3D structure, and (2) point cloud at the previous timestep should be reconstructable using reverse motion from future timestep. Extensive experiments show that FastPCI significantly outperforms the state-of-the-art PointINet and NeuralPCI with notable gains (e.g. 26.6% and 18.3% reduction in Chamfer Distance in KITTI), while being more than 10x and 600x faster, respectively. Code is available at https://github.com/genuszty/FastPCI
title FastPCI: Motion-Structure Guided Fast Point Cloud Frame Interpolation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.19573