ANFPCGC++: Point Cloud Geometry Coding Using Augmented Normalizing Flows and Transformer-Based Entropy Model

As immersive media gains increasing prominence, point clouds have emerged as a preferred data representation for presenting complex 3D scenes. However, the large size of point cloud data poses challenges in terms of storage and real-time transmission, prompting the need for highly efficient point cl...

Full description

Saved in:
Bibliographic Details
Main Authors: Jui-Chiu Chiang, Ji-Jin Chiu, Monyneath Yim
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10735187/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846170452760723456
author Jui-Chiu Chiang
Ji-Jin Chiu
Monyneath Yim
author_facet Jui-Chiu Chiang
Ji-Jin Chiu
Monyneath Yim
author_sort Jui-Chiu Chiang
collection DOAJ
description As immersive media gains increasing prominence, point clouds have emerged as a preferred data representation for presenting complex 3D scenes. However, the large size of point cloud data poses challenges in terms of storage and real-time transmission, prompting the need for highly efficient point cloud compression techniques. In response to these challenges, we introduce a novel approach called ANFPCGC++ (Augmented Normalizing Flow-based Point Cloud Geometry Compression) for lossy static point cloud geometry coding. ANFPCGC++ leverages the power of Augmented Normalizing Flow (ANF) in conjunction with sparse convolution to effectively capture and incorporate spatial correlations inherent in point clouds. ANF offers a higher level of expressiveness compared to conventional methods like variational autoencoders (VAE), resulting in more accurate and faithful latent representations. Furthermore, we introduce a Transformer-based entropy model, that combines the hyperprior and context information, enabling a more precise entropy model that supports parallel computation. Extensive experimental results confirm the superior performance of ANFPCGC++. By comparing to the point cloud coding standards G-PCC and V-PCC, our proposed method achieves remarkable bitrate savings of 63.7% and 60.0% in terms of D1-PSNR, respectively. Additionally, when compared to other deep learning-based point cloud geometry compression methods like PCGCv2 and ANFPCGC, our approach demonstrates an average bitrate reduction of 25.6% and 23.6% in terms of D1-PSNR, respectively. The source code is available at <uri>https://github.com/ymnn1996/ANFPCGC2</uri>.
format Article
id doaj-art-9732597d5e33415980d35f2a5af49ad8
institution Kabale University
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-9732597d5e33415980d35f2a5af49ad82024-11-12T00:01:32ZengIEEEIEEE Access2169-35362024-01-011216341016342310.1109/ACCESS.2024.348646410735187ANFPCGC++: Point Cloud Geometry Coding Using Augmented Normalizing Flows and Transformer-Based Entropy ModelJui-Chiu Chiang0https://orcid.org/0000-0003-1397-8393Ji-Jin Chiu1Monyneath Yim2https://orcid.org/0009-0005-9402-4744Department of Electrical Engineering, National Chung Cheng University, Chiayi, TaiwanDepartment of Electrical Engineering, National Chung Cheng University, Chiayi, TaiwanDepartment of Electrical Engineering, National Chung Cheng University, Chiayi, TaiwanAs immersive media gains increasing prominence, point clouds have emerged as a preferred data representation for presenting complex 3D scenes. However, the large size of point cloud data poses challenges in terms of storage and real-time transmission, prompting the need for highly efficient point cloud compression techniques. In response to these challenges, we introduce a novel approach called ANFPCGC++ (Augmented Normalizing Flow-based Point Cloud Geometry Compression) for lossy static point cloud geometry coding. ANFPCGC++ leverages the power of Augmented Normalizing Flow (ANF) in conjunction with sparse convolution to effectively capture and incorporate spatial correlations inherent in point clouds. ANF offers a higher level of expressiveness compared to conventional methods like variational autoencoders (VAE), resulting in more accurate and faithful latent representations. Furthermore, we introduce a Transformer-based entropy model, that combines the hyperprior and context information, enabling a more precise entropy model that supports parallel computation. Extensive experimental results confirm the superior performance of ANFPCGC++. By comparing to the point cloud coding standards G-PCC and V-PCC, our proposed method achieves remarkable bitrate savings of 63.7% and 60.0% in terms of D1-PSNR, respectively. Additionally, when compared to other deep learning-based point cloud geometry compression methods like PCGCv2 and ANFPCGC, our approach demonstrates an average bitrate reduction of 25.6% and 23.6% in terms of D1-PSNR, respectively. The source code is available at <uri>https://github.com/ymnn1996/ANFPCGC2</uri>.https://ieeexplore.ieee.org/document/10735187/Point cloud geometry compressionaugmented normalizing flowcontext modeltransformer
spellingShingle Jui-Chiu Chiang
Ji-Jin Chiu
Monyneath Yim
ANFPCGC++: Point Cloud Geometry Coding Using Augmented Normalizing Flows and Transformer-Based Entropy Model
IEEE Access
Point cloud geometry compression
augmented normalizing flow
context model
transformer
title ANFPCGC++: Point Cloud Geometry Coding Using Augmented Normalizing Flows and Transformer-Based Entropy Model
title_full ANFPCGC++: Point Cloud Geometry Coding Using Augmented Normalizing Flows and Transformer-Based Entropy Model
title_fullStr ANFPCGC++: Point Cloud Geometry Coding Using Augmented Normalizing Flows and Transformer-Based Entropy Model
title_full_unstemmed ANFPCGC++: Point Cloud Geometry Coding Using Augmented Normalizing Flows and Transformer-Based Entropy Model
title_short ANFPCGC++: Point Cloud Geometry Coding Using Augmented Normalizing Flows and Transformer-Based Entropy Model
title_sort anfpcgc point cloud geometry coding using augmented normalizing flows and transformer based entropy model
topic Point cloud geometry compression
augmented normalizing flow
context model
transformer
url https://ieeexplore.ieee.org/document/10735187/
work_keys_str_mv AT juichiuchiang anfpcgcpointcloudgeometrycodingusingaugmentednormalizingflowsandtransformerbasedentropymodel
AT jijinchiu anfpcgcpointcloudgeometrycodingusingaugmentednormalizingflowsandtransformerbasedentropymodel
AT monyneathyim anfpcgcpointcloudgeometrycodingusingaugmentednormalizingflowsandtransformerbasedentropymodel