Video Instance Segmentation 2019: A winning approach for combined Detection, Segmentation, Classification and Tracking

被引：11
|
作者：
Luiten, Jonathon [1 ,2 ]
Torr, Philip H. S. [2 ]
Leibe, Bastian [1 ]
机构：
[1] Rhein Westfal TH Aachen, Aachen, Germany
[2] Univ Oxford, Oxford, England
来源：
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW) | 2019年
基金：
英国工程与自然科学研究理事会;
关键词：
D O I：
10.1109/ICCVW.2019.00088
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Video Instance Segmentation (VIS) is the task of localizing all objects in a video, segmenting them, tracking them throughout the video and classifying them into a set of pre-defined classes. In this work, divide VIS into these four parts: detection, segmentation, tracking and. classification. We then develop algorithms for petforming each of these four sub tasks individually, and combine these into a complete solution for VIS. Our solution is an adaptation of Un-OVOST, the current best petforming algorithm for Unsupervised Video Object Segmentation, to this VIS task. We benchmark our algorithm on the 2019 YouTube-VIS Challenge, where we obtain first place with an mAP score of 46.7%.
引用
收藏
页码：709 / 712
页数：4
相关论文
共 50 条
  • [1] Video Instance Segmentation
    Yang, Linjie
    Fan, Yuchen
    Xu, Ning
    2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, : 5187 - 5196
  • [2] Occluded Video Instance Segmentation with Set Prediction Approach
    Bae, Heechul
    Song, Soonyong
    Park, Junhee
    2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW 2021), 2021, : 3843 - 3846
  • [3] Adapting Video Instance Segmentation for Instance Search
    Nguyen, An Thi
    20TH INTERNATIONAL CONFERENCE ON CONTENT-BASED MULTIMEDIA INDEXING, CBMI 2023, 2023, : 256 - 260
  • [4] Video Instance Segmentation by Instance Flow Assembly
    Li, Xiang
    Wang, Jinglu
    Li, Xiao
    Lu, Yan
    IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 : 7469 - 7479
  • [5] CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks
    Casado-Garcia, Angela
    Dominguez, Cesar
    Garcia-Dominguez, Manuel
    Heras, Jonathan
    Ines, Adrian
    Mata, Eloy
    Pascual, Vico
    BMC BIOINFORMATICS, 2019, 20 (1)
  • [6] CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks
    Ángela Casado-García
    César Domínguez
    Manuel García-Domínguez
    Jónathan Heras
    Adrián Inés
    Eloy Mata
    Vico Pascual
    BMC Bioinformatics, 20
  • [7] An Empirical Study of Detection-Based Video Instance Segmentation
    Wang, Qiang
    He, Yi
    Yang, Xiaoyun
    Yang, Zhao
    Torr, Philip H. S.
    2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW), 2019, : 713 - 716
  • [8] Integration of regularized l1 tracking and instance segmentation for video object tracking
    Gurkan, Filiz
    Gunsel, Bilge
    NEUROCOMPUTING, 2021, 423 : 284 - 300
  • [9] Video object segmentation and tracking using ψ-learning classification
    Liu, Y
    Zheng, YF
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2005, 15 (07) : 885 - 899
  • [10] Occluded Video Instance Segmentation: A Benchmark
    Jiyang Qi
    Yan Gao
    Yao Hu
    Xinggang Wang
    Xiaoyu Liu
    Xiang Bai
    Serge Belongie
    Alan Yuille
    Philip H. S. Torr
    Song Bai
    International Journal of Computer Vision, 2022, 130 : 2022 - 2039
12345