Panorama: A Data System for Unbounded Vocabulary Querying over Video

被引:17
|
作者
Zhang, Yuhao [1 ]
Kumar, Arun [1 ]
机构
[1] Univ Calif San Diego, San Diego, CA 92103 USA
来源
PROCEEDINGS OF THE VLDB ENDOWMENT | 2019年 / 13卷 / 04期
关键词
D O I
10.14778/3372716.3372721
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Deep convolutional neural networks (CNNs) achieve state-of-the-art accuracy for many computer vision tasks. But using them for video monitoring applications incurs high computational cost and inference latency. Thus, recent works have studied how to improve system efficiency. But they largely focus on small "closed world" prediction vocabularies even though many applications in surveillance security, traffic analytics, etc. have an ever-growing set of target entities. We call this the "unbounded vocabulary" issue, and it is a key bottleneck for emerging video monitoring applications. We present the first data system for tacking this issue for video querying, Panorama. Our design philosophy is to build a unified and domain-agnostic system that lets application users generalize to unbounded vocabularies in an out-of-the-box manner without tedious manual re-training. To this end, we synthesize and innovate upon an array of techniques from the ML, vision, databases, and multimedia systems literature to devise a new system architecture. We also present techniques to ensure Panorama has high inference efficiency. Experiments with multiple real-world datasets show that Panorama can achieve between 2x to 20x higher efficiency than baseline approaches on in-vocabulary queries, while still yielding comparable accuracy and also generalizing well to unbounded vocabularies.
引用
收藏
页码:477 / 491
页数:15
相关论文
共 50 条
  • [1] An Affect-Based Video Retrieval System with Open Vocabulary Querying
    Chan, Ching Hau
    Jones, Gareth J. F.
    [J]. ADAPTIVE MULTIMEDIA RETRIEVAL: CONTEXT, EXPLORATION, AND FUSION, 2012, 6817 : 103 - 117
  • [2] Panorama video server system
    Okimura, T
    Kimura, K
    Nakazawa, K
    Nakajima, H
    [J]. STEREOSCOPIC DISPLAYS AND VIRTUAL REALITY SYSTEMS V, 1998, 3295 : 381 - 390
  • [3] Database querying with personalized vocabulary using data summaries
    Ughetto, L.
    Voglozin, W. A.
    Mouaddib, N.
    [J]. FUZZY SETS AND SYSTEMS, 2008, 159 (15) : 2030 - 2046
  • [4] A multi-paradigm video querying system over the web - Architecture and mechanisms
    Chan, SSM
    Li, Q
    [J]. COOPERATIVE INTERNET COMPUTING, 2003, 729 : 37 - 50
  • [5] Querying XML data over DHT system using XPeer
    Rao, WX
    Song, H
    Ma, FY
    [J]. GRID AND COOPERATIVE COMPUTING GCC 2004, PROCEEDINGS, 2004, 3251 : 559 - 566
  • [6] Interactive Panorama Video Distribution System
    Kimata, Hideaki
    Isogai, Megumi
    Noto, Hajime
    Inoue, Masayuki
    Fukazawa, Katsuhiko
    Matsuura, Norihiko
    [J]. 2011 TECHNICAL SYMPOSIUM AT ITU TELECOM WORLD (ITU WT), 2011, : 45 - 50
  • [7] Modeling and querying video data: A hybrid approach
    Decleir, C
    Hacid, MS
    Kouloumdjian, J
    [J]. IEEE WORKSHOP ON CONTENT-BASED ACCESS OF IMAGE AND VIDEO LIBRARIES - PROCEEDINGS, 1998, : 86 - 90
  • [8] Video data modeling and querying on surveillance videos
    Durak, Nurcan
    Yazici, Adnan
    [J]. 2006 IEEE 14TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS, VOLS 1 AND 2, 2006, : 976 - +
  • [9] A database approach for modeling and querying video data
    Decleir, C
    Hacid, MS
    Kouloumdjian, J
    [J]. 15TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING, PROCEEDINGS, 1999, : 6 - 13
  • [10] A database approach for modeling and querying video data
    Hacid, MS
    Decleir, C
    Kouloumdjian, J
    [J]. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2000, 12 (05) : 729 - 750