Boosting neural video codecs by exploiting hierarchical redundancy

被引:3
|
作者
Pourreza, Reza [1 ]
Le, Hoang [1 ]
Said, Amir [1 ]
Sautiere, Guillaume [1 ]
Wiggers, Auke [1 ]
机构
[1] Qualcomm AI Res, San Diego, CA 92121 USA
关键词
D O I
10.1109/WACV56688.2023.00532
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In video compression, coding efficiency is improved by reusing pixels from previously decoded frames via motion and residual compensation. We define two levels of hierarchical redundancy in video frames: 1) first-order: redundancy in pixel space, i.e., similarities in pixel values across neighboring frames, which is effectively captured using motion and residual compensation, 2) second-order: redundancy in motion and residual maps due to smooth motion in natural videos. While most of the existing neural video coding literature addresses first-order redundancy, we tackle the problem of capturing second-order redundancy in neural video codecs via predictors. We introduce generic motion and residual predictors that learn to extrapolate from previously decoded data. These predictors are lightweight, and can be employed with most neural video codecs in order to improve their rate-distortion performance. Moreover, while RGB is the dominant colorspace in neural video coding literature, we introduce general modifications for neural video codecs to embrace the YUV420 colorspace and report YUV420 results. Our experiments show that using our predictors with a well-known neural video codec leads to 38% and 34% bitrate savings in RGB and YUV420 colorspaces measured on the UVG dataset.
引用
收藏
页码:5344 / 5353
页数:10
相关论文
共 50 条
  • [1] Exploiting Latent Properties to Optimize Neural Codecs
    Balcilar, Muhammet
    Damodaran, Bharath Bhushan
    Naser, Karam
    Galpin, Franck
    Hellier, Pierre
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2025, 34 : 306 - 319
  • [2] Swift: Adaptive Video Streaming with Layered Neural Codecs
    Dasari, Mallesham
    Kahatapitiya, Kumara
    Das, Samir R.
    Balasubramanian, Aruna
    Samaras, Dimitris
    PROCEEDINGS OF THE 19TH USENIX SYMPOSIUM ON NETWORKED SYSTEMS DESIGN AND IMPLEMENTATION (NSDI '22), 2022, : 103 - 118
  • [3] Exploiting Temporal Redundancy of Visual Structures for Video Compression
    Georgiadis, Georgios
    Soatto, Stefano
    2015 DATA COMPRESSION CONFERENCE (DCC), 2015, : 445 - 445
  • [4] Robust Video Decoding by Exploiting Residual Source Redundancy
    Sun, Lulu
    Gao, Shaoshuai
    2015 INTERNATIONAL CONFERENCE ON WIRELESS COMMUNICATIONS & SIGNAL PROCESSING (WCSP), 2015,
  • [5] Performance analysis of hierarchical transform coding with a large kernel for video codecs
    Lee, Bumshik
    Kim, Munchurl
    Kim, Hui Yong
    Choi, Jin Soo
    IET IMAGE PROCESSING, 2014, 8 (01) : 12 - 22
  • [6] Exploiting global redundancy in big surveillance video data for efficient coding
    Xiao, Jing
    Liao, Liang
    Hu, Jinhui
    Chen, Yu
    Hu, Ruimin
    CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS, 2015, 18 (02): : 531 - 540
  • [7] Exploiting global redundancy in big surveillance video data for efficient coding
    Jing Xiao
    Liang Liao
    Jinhui Hu
    Yu Chen
    Ruimin Hu
    Cluster Computing, 2015, 18 : 531 - 540
  • [8] Boosting the Hierarchical Hyperellipsoidal Neural Gas Networks
    Fang, Xiufen
    Liu, Guisong
    Huang, Tingzhu
    ICNC 2008: FOURTH INTERNATIONAL CONFERENCE ON NATURAL COMPUTATION, VOL 3, PROCEEDINGS, 2008, : 126 - +
  • [9] Exploiting Non-Linear Redundancy for Neural Model Compression
    Shah, Muhammad A.
    Olivier, Raphael
    Raj, Bhiksha
    2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2021, : 9928 - 9935
  • [10] Comparative Analysis: Conventional Video Codecs v/s Compressive Sensing Video Codecs
    Ebrahim, Mansoor
    Adil, Syed Hasan
    Gul, Tayyab
    Raza, Kamran
    2018 3RD INTERNATIONAL CONFERENCE ON EMERGING TRENDS IN ENGINEERING, SCIENCES AND TECHNOLOGY (ICEEST), 2018,