HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation

被引:3
|
作者
Zhou, Zijian [1 ]
Shi, Miaojing [2 ]
Caesar, Holger [3 ]
机构
[1] Kings Coll London, Dept Informat, London, England
[2] Tongji Univ, Coll Elect & Informat Engn, Shanghai, Peoples R China
[3] Delft Univ Technol, Intelligent Vehicles Lab, Delft, Netherlands
关键词
D O I
10.1109/ICCV51070.2023.01978
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Panoptic Scene Graph generation (PSG) is a recently proposed task in image scene understanding that aims to segment the image and extract triplets of subjects, objects and their relations to build a scene graph. This task is particularly challenging for two reasons. First, it suffers from a long-tail problem in its relation categories, making naive biased methods more inclined to high-frequency relations. Existing unbiased methods tackle the long-tail problem by data/loss rebalancing to favor low-frequency relations. Second, a subject-object pair can have two or more semantically overlapping relations. While existing methods favor one over the other, our proposed HiLo framework lets different network branches specialize on low and high frequency relations, enforce their consistency and fuse the results. To the best of our knowledge we are the first to propose an explicitly unbiased PSG method. In extensive experiments we show that our HiLo framework achieves state-of-the-art results on the PSG task. We also apply our method to the Scene Graph Generation task that predicts boxes instead of masks and see improvements over all baseline methods. Code is available at https://github.com/franciszzj/HiLo.
引用
收藏
页码:21580 / 21591
页数:12
相关论文
共 50 条
  • [21] Label Semantic Knowledge Distillation for Unbiased Scene Graph Generation
    Li, Lin
    Xiao, Jun
    Shi, Hanrong
    Wang, Wenxiao
    Shao, Jian
    Liu, An-An
    Yang, Yi
    Chen, Long
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2024, 34 (01) : 195 - 206
  • [22] Fast Contextual Scene Graph Generation with Unbiased Context Augmentation
    Jin, Tianlei
    Guo, Fangtai
    Meng, Qiwei
    Zhu, Shiqiang
    Xi, Xiangming
    Wang, Wen
    Mu, Zonghao
    Song, Wei
    2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR, 2023, : 6302 - 6311
  • [23] Divide-and-Conquer Predictor for Unbiased Scene Graph Generation
    Han, Xianjing
    Dong, Xingning
    Song, Xuemeng
    Gan, Tian
    Zhan, Yibing
    Yan, Yan
    Nie, Liqiang
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2022, 32 (12) : 8611 - 8622
  • [24] An Effective Dynamic Reweighting Method for Unbiased Scene Graph Generation
    Hu, Lingfeng
    Liu, Si
    Wang, Hanzi
    PATTERN RECOGNITION AND COMPUTER VISION, PRCV 2023, PT I, 2024, 14425 : 345 - 356
  • [25] CogTree: Cognition Tree Loss for Unbiased Scene Graph Generation
    Yu, Jing
    Chai, Yuan
    Wang, Yujing
    Hu, Yue
    Wu, Qi
    PROCEEDINGS OF THE THIRTIETH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2021, 2021, : 1274 - 1280
  • [26] Bipartite Graph Network with Adaptive Message Passing for Unbiased Scene Graph Generation
    Li, Rongjie
    Zhang, Songyang
    Wan, Bo
    He, Xuming
    2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 11104 - 11114
  • [27] Unbiased scene graph generation using the self-distillation method
    Sun, Bo
    Hao, Zhuo
    Yu, Lejun
    He, Jun
    VISUAL COMPUTER, 2024, 40 (04): : 2381 - 2390
  • [28] TEMPLATE-GUIDED DATA AUGMENTATION FOR UNBIASED SCENE GRAPH GENERATION
    Zang, Yujie
    Li, Yaochen
    Cao, Luguang
    Lu, Ruitao
    2024 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, ICASSP 2024, 2024, : 3510 - 3514
  • [29] Knowledge-Enhanced Context Representation for Unbiased Scene Graph Generation
    Wang, Yuanlong
    Liu, Zhenqi
    Zhang, Hu
    Li, Ru
    WEB AND BIG DATA, APWEB-WAIM 2024, PT I, 2024, 14961 : 248 - 263
  • [30] Unbiased scene graph generation using the self-distillation method
    Bo Sun
    Zhuo Hao
    Lejun Yu
    Jun He
    The Visual Computer, 2024, 40 : 2381 - 2390