AdaPoinTr: Diverse Point Cloud Completion With Adaptive Geometry-Aware Transformers
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2023-12 · 45 (12) · 14114-14130
DOI: 10.1109/tpami.2023.3309253 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023-12. Verfasst von Xumin Yu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 170-mal zitiert (Crossref) und nennt 84 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
In this paper, we propose a Transformer encoder-decoder architecture, called PoinTr, which reformulates point cloud completion as a set-to-set translation problem and employs a geometry-aware block to model local geometric relationships explicitly. The migration of Transformers enables our model to better learn structural knowledge and preserve detailed information for point cloud completion. Taking a step towards more complicated and diverse situations, we further propose AdaPoinTr by developing an adaptive query generation mechanism and designing a novel denoising task during completing a point cloud. Coupling these two techniques enables us to train the model efficiently and effectively: we reduce training time (by 15x or more) and improve completion performance (over 20%). Additionally, we propose two more challenging benchmarks with more diverse incomplete point clouds that can better reflect real-world scenarios to promote future research. We also show our method can be extended to the scene-level point cloud completion scenario by designing a new geometry-enhanced semantic scene completion framework. Extensive experiments on the existing and newly-proposed datasets demonstrate the effectiveness of our method, which attains 6.53 CD on PCN, 0.81 CD on ShapeNet-55 and 0.392 MMD on real-world KITTI, surpassing other work by a large margin and establishing new state-of-the-arts on various benchmarks. Most notably, AdaPoinTr can achieve such promising performance with higher throughputs and fewer FLOPs compared with the previous best methods in practice.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Yu, X., Rao, Y., Wang, Z., Lu, J., & Zhou, J. (2023). AdaPoinTr: Diverse Point Cloud Completion With Adaptive Geometry-Aware Transformers. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(12), 14114-14130. https://doi.org/10.1109/tpami.2023.3309253
Vancouver
Yu X, Rao Y, Wang Z, Lu J, Zhou J. AdaPoinTr: Diverse Point Cloud Completion With Adaptive Geometry-Aware Transformers. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2023;45(12):14114-14130. doi:10.1109/tpami.2023.3309253
BibTeX
@article{Yu2023,
author = {Yu, Xumin and Rao, Yongming and Wang, Ziyi and Lu, Jiwen and Zhou, Jie},
title = {AdaPoinTr: Diverse Point Cloud Completion With Adaptive Geometry-Aware Transformers},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2023},
volume = {45},
number = {12},
pages = {14114--14130},
doi = {10.1109/tpami.2023.3309253}
}
EUVIMED