Face Recognition on Point Cloud with Cgan-Top for Denoising

Junyu Liu; Jianfeng Ren; Hongliang Sun; Xudong Jiang

doi:10.1109/ICASSP49357.2023.10096203

Face Recognition on Point Cloud with Cgan-Top for Denoising

Junyu Liu, Jianfeng Ren, Hongliang Sun, Xudong Jiang

School of Computer Science

Research output: Journal Publication › Conference article › peer-review

1 Citation (Scopus)

Abstract

Face recognition using 3D point clouds is gaining growing interest, while raw point clouds often contain a significant amount of noise due to imperfect sensors. In this paper, an end-to-end 3D face recognition on a noisy point cloud is proposed, which synergistically integrates the denoising and recognition modules. Specifically, a Conditional Generative Adversarial Network on Three Orthogonal Planes (cGAN-TOP) is designed to effectively remove the noise in the point cloud, and recover the underlying features for subsequent recognition. A Linked Dynamic Graph Convolutional Neural Network (LDGCNN) is then adapted to recognize faces from the processed point cloud, which hierarchically links both the local point features and neighboring features of multiple scales. The proposed method is validated on the Bosphorus dataset. It significantly improves the recognition accuracy under all noise settings, with a maximum gain of 14.81%.

Original language	English
Journal	Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing
DOIs	https://doi.org/10.1109/ICASSP49357.2023.10096203
Publication status	Published - 2023
Event	48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023 - Rhodes Island, Greece Duration: 4 Jun 2023 → 10 Jun 2023

Keywords

3D Point Cloud
Conditional Generative Neural Network on Three Orthogonal Planes
Face Recognition
Linked Dynamic Graph Convolutional Neural Network

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP49357.2023.10096203

Cite this

@article{0541a013ae784c00ae2cfe49ff5a7d60,

title = "Face Recognition on Point Cloud with Cgan-Top for Denoising",

abstract = "Face recognition using 3D point clouds is gaining growing interest, while raw point clouds often contain a significant amount of noise due to imperfect sensors. In this paper, an end-to-end 3D face recognition on a noisy point cloud is proposed, which synergistically integrates the denoising and recognition modules. Specifically, a Conditional Generative Adversarial Network on Three Orthogonal Planes (cGAN-TOP) is designed to effectively remove the noise in the point cloud, and recover the underlying features for subsequent recognition. A Linked Dynamic Graph Convolutional Neural Network (LDGCNN) is then adapted to recognize faces from the processed point cloud, which hierarchically links both the local point features and neighboring features of multiple scales. The proposed method is validated on the Bosphorus dataset. It significantly improves the recognition accuracy under all noise settings, with a maximum gain of 14.81%.",

keywords = "3D Point Cloud, Conditional Generative Neural Network on Three Orthogonal Planes, Face Recognition, Linked Dynamic Graph Convolutional Neural Network",

author = "Junyu Liu and Jianfeng Ren and Hongliang Sun and Xudong Jiang",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023 ; Conference date: 04-06-2023 Through 10-06-2023",

year = "2023",

doi = "10.1109/ICASSP49357.2023.10096203",

language = "English",

journal = "Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing",

issn = "1520-6149",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Face Recognition on Point Cloud with Cgan-Top for Denoising

AU - Liu, Junyu

AU - Ren, Jianfeng

AU - Sun, Hongliang

AU - Jiang, Xudong

PY - 2023

Y1 - 2023

N2 - Face recognition using 3D point clouds is gaining growing interest, while raw point clouds often contain a significant amount of noise due to imperfect sensors. In this paper, an end-to-end 3D face recognition on a noisy point cloud is proposed, which synergistically integrates the denoising and recognition modules. Specifically, a Conditional Generative Adversarial Network on Three Orthogonal Planes (cGAN-TOP) is designed to effectively remove the noise in the point cloud, and recover the underlying features for subsequent recognition. A Linked Dynamic Graph Convolutional Neural Network (LDGCNN) is then adapted to recognize faces from the processed point cloud, which hierarchically links both the local point features and neighboring features of multiple scales. The proposed method is validated on the Bosphorus dataset. It significantly improves the recognition accuracy under all noise settings, with a maximum gain of 14.81%.

AB - Face recognition using 3D point clouds is gaining growing interest, while raw point clouds often contain a significant amount of noise due to imperfect sensors. In this paper, an end-to-end 3D face recognition on a noisy point cloud is proposed, which synergistically integrates the denoising and recognition modules. Specifically, a Conditional Generative Adversarial Network on Three Orthogonal Planes (cGAN-TOP) is designed to effectively remove the noise in the point cloud, and recover the underlying features for subsequent recognition. A Linked Dynamic Graph Convolutional Neural Network (LDGCNN) is then adapted to recognize faces from the processed point cloud, which hierarchically links both the local point features and neighboring features of multiple scales. The proposed method is validated on the Bosphorus dataset. It significantly improves the recognition accuracy under all noise settings, with a maximum gain of 14.81%.

KW - 3D Point Cloud

KW - Conditional Generative Neural Network on Three Orthogonal Planes

KW - Face Recognition

KW - Linked Dynamic Graph Convolutional Neural Network

UR - http://www.scopus.com/inward/record.url?scp=85180571988&partnerID=8YFLogxK

U2 - 10.1109/ICASSP49357.2023.10096203

DO - 10.1109/ICASSP49357.2023.10096203

M3 - Conference article

AN - SCOPUS:85180571988

SN - 1520-6149

JO - Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

JF - Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

T2 - 48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

Y2 - 4 June 2023 through 10 June 2023

ER -

Face Recognition on Point Cloud with Cgan-Top for Denoising

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this