RSCA: Real-time Segmentation-based Context-Aware Scene Text Detection

被引:11
|
作者
Li, Jiachen [1 ,3 ]
Lin, Yuan [3 ]
Liu, Rongrong [3 ]
Ho, Chiu Man [3 ]
Shi, Humphrey [1 ,2 ]
机构
[1] UIUC, Urbana, IL 61801 USA
[2] Univ Oregon, Eugene, OR 97403 USA
[3] InnoPeak Technol, Palo Alto, CA 94303 USA
关键词
D O I
10.1109/CVPRW53098.2021.00267
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Segmentation-based scene text detection methods have been widely adopted for arbitrary-shaped text detection recently, since they make accurate pixel-level predictions on curved text instances and can facilitate real-time inference without time-consuming processing on anchors. However, current segmentation-based models are unable to learn the shapes of curved texts and often require complex label assignments or repeated feature aggregations for more accurate detection. In this paper, we propose RSCA: a Real-time Segmentation-based Context-Aware model for arbitrary-shaped scene text detection, which sets a strong baseline for scene text detection with two simple yet effective strategies: Local Context-Aware Upsampling and Dynamic Text-Spine Labeling, which model local spatial transformation and simplify label assignments separately. Based on these strategies, RSCA achieves state-of-the-art performance in both speed and accuracy, without complex label assignments or repeated feature aggregations. We conduct extensive experiments on multiple benchmarks to validate the effectiveness of our method. RSCA-640 reaches 83.9% F-measure at 48.3 FPS on CTW1500 dataset.
引用
收藏
页码:2349 / 2358
页数:10
相关论文
共 50 条
  • [1] Segmentation-based context-aware enhancement network for medical images
    Bao, Hua
    Li, Qing
    Zhu, Yuqing
    INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2024, 15 (03) : 963 - 983
  • [2] Segmentation-based context-aware enhancement network for medical images
    Hua Bao
    Qing Li
    Yuqing Zhu
    International Journal of Machine Learning and Cybernetics, 2024, 15 : 963 - 983
  • [3] PolypSeg plus : A Lightweight Context-Aware Network for Real-Time Polyp Segmentation
    Wu, Huisi
    Zhao, Zebin
    Zhong, Jiafu
    Wang, Wei
    Wen, Zhenkun
    Qin, Jing
    IEEE TRANSACTIONS ON CYBERNETICS, 2023, 53 (04) : 2610 - 2621
  • [4] An Accurate Segmentation-Based Scene Text Detector with Context Attention and Repulsive Text Border
    Liu, Xi
    Zhou, Gaojing
    Zhang, Rui
    Wei, Xiaolin
    2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2020), 2020, : 2344 - 2352
  • [5] Real-time Scene Text Detection Based on Stroke Model
    Liu, Yi
    Zhang, Dongming
    Zhang, Yongdong
    Lin, Shouxun
    2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2014, : 3116 - 3120
  • [6] Context-Aware Adaptive Weighted Attention Network for Real-Time Surface Defect Segmentation
    Zhang, Gaowei
    Lu, Yang
    Jiang, Xiaoheng
    Yan, Feng
    Xu, Mingliang
    IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2024, 73
  • [7] CDText: Scene text detector based on context-aware deformable transformer
    Wu, Yirui
    Kong, Qiran
    Yong, Lai
    Narducci, Fabio
    Wan, Shaohua
    PATTERN RECOGNITION LETTERS, 2023, 172 : 8 - 14
  • [8] Real-time context-aware social media recommendation
    Zhou, Xiangmin
    Qin, Dong
    Chen, Lei
    Zhang, Yanchun
    VLDB JOURNAL, 2019, 28 (02): : 197 - 219
  • [9] Real-time context-aware social media recommendation
    Xiangmin Zhou
    Dong Qin
    Lei Chen
    Yanchun Zhang
    The VLDB Journal, 2019, 28 : 197 - 219
  • [10] Context-aware Real-time Population Estimation for Metropolis
    Xu, Fengli
    Feng, Jie
    Zhang, Pengyu
    Li, Yong
    UBICOMP'16: PROCEEDINGS OF THE 2016 ACM INTERNATIONAL JOINT CONFERENCE ON PERVASIVE AND UBIQUITOUS COMPUTING, 2016, : 1064 - 1075