The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Pemampatan awan titik berasaskan video (V-PCC) menggunakan teknologi pemampatan video untuk mengekod awan titik padat dengan cekap memberikan prestasi pemampatan terkini dengan beban pengiraan yang agak kecil. V-PCC menukar data awan titik 3 dimensi kepada tiga jenis bingkai 2 dimensi, iaitu, penghunian, geometri dan bingkai atribut, dan mengekodnya melalui pemampatan video. Sebaliknya, kualiti bingkai ini mungkin merosot disebabkan oleh pemampatan video. Kertas kerja ini mencadangkan penapis pasca pemprosesan berasaskan rangkaian saraf adaptif pada bingkai atribut untuk mengurangkan masalah degradasi. Tambahan pula, kaedah latihan baru menggunakan bingkai penghunian dikaji. Keputusan eksperimen menunjukkan purata keuntungan kadar BD sebanyak 3.0%, 29.3% dan 22.2% masing-masing untuk Y, U dan V.
Keiichiro TAKADA
Sharp Corporation
Yasuaki TOKUMO
Sharp Corporation
Tomohiro IKAI
Sharp Corporation
Takeshi CHUJOH
Sharp Corporation
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Keiichiro TAKADA, Yasuaki TOKUMO, Tomohiro IKAI, Takeshi CHUJOH, "Neural Network-Based Post-Processing Filter on V-PCC Attribute Frames" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 10, pp. 1673-1676, October 2023, doi: 10.1587/transinf.2023PCL0002.
Abstract: Video-based point cloud compression (V-PCC) utilizes video compression technology to efficiently encode dense point clouds providing state-of-the-art compression performance with a relatively small computation burden. V-PCC converts 3-dimensional point cloud data into three types of 2-dimensional frames, i.e., occupancy, geometry, and attribute frames, and encodes them via video compression. On the other hand, the quality of these frames may be degraded due to video compression. This paper proposes an adaptive neural network-based post-processing filter on attribute frames to alleviate the degradation problem. Furthermore, a novel training method using occupancy frames is studied. The experimental results show average BD-rate gains of 3.0%, 29.3% and 22.2% for Y, U and V respectively.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2023PCL0002/_p
Salinan
@ARTICLE{e106-d_10_1673,
author={Keiichiro TAKADA, Yasuaki TOKUMO, Tomohiro IKAI, Takeshi CHUJOH, },
journal={IEICE TRANSACTIONS on Information},
title={Neural Network-Based Post-Processing Filter on V-PCC Attribute Frames},
year={2023},
volume={E106-D},
number={10},
pages={1673-1676},
abstract={Video-based point cloud compression (V-PCC) utilizes video compression technology to efficiently encode dense point clouds providing state-of-the-art compression performance with a relatively small computation burden. V-PCC converts 3-dimensional point cloud data into three types of 2-dimensional frames, i.e., occupancy, geometry, and attribute frames, and encodes them via video compression. On the other hand, the quality of these frames may be degraded due to video compression. This paper proposes an adaptive neural network-based post-processing filter on attribute frames to alleviate the degradation problem. Furthermore, a novel training method using occupancy frames is studied. The experimental results show average BD-rate gains of 3.0%, 29.3% and 22.2% for Y, U and V respectively.},
keywords={},
doi={10.1587/transinf.2023PCL0002},
ISSN={1745-1361},
month={October},}
Salinan
TY - JOUR
TI - Neural Network-Based Post-Processing Filter on V-PCC Attribute Frames
T2 - IEICE TRANSACTIONS on Information
SP - 1673
EP - 1676
AU - Keiichiro TAKADA
AU - Yasuaki TOKUMO
AU - Tomohiro IKAI
AU - Takeshi CHUJOH
PY - 2023
DO - 10.1587/transinf.2023PCL0002
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2023
AB - Video-based point cloud compression (V-PCC) utilizes video compression technology to efficiently encode dense point clouds providing state-of-the-art compression performance with a relatively small computation burden. V-PCC converts 3-dimensional point cloud data into three types of 2-dimensional frames, i.e., occupancy, geometry, and attribute frames, and encodes them via video compression. On the other hand, the quality of these frames may be degraded due to video compression. This paper proposes an adaptive neural network-based post-processing filter on attribute frames to alleviate the degradation problem. Furthermore, a novel training method using occupancy frames is studied. The experimental results show average BD-rate gains of 3.0%, 29.3% and 22.2% for Y, U and V respectively.
ER -