C Yang, XC Yin, WY Pei, S Tian, ZY Zuo… - … on Image Processing, 2017 - ieeexplore.ieee.org
… • Different from most conventional video textdetection methods using … based scenetext detection approach using multiple frames, which combines detection, tracking and recognition by …
… detection robustness from our observation. To address these challenges, this paper proposes a concise DynamicPointTextDEtection … directly leverages explicit point coordinates to gen…
J Li, Y Lin, R Liu, CM Ho, H Shi - … and pattern recognition, 2021 - openaccess.thecvf.com
… point refers to non-textregion prediction and bright red point … -shaped scenetextdetection, we propose a dynamictext-… and assignment, which dynamically enlarges text-spine as labels …
J Zheng, L Zhang, Y Wu, C Zhao - ICASSP 2024-2024 IEEE …, 2024 - ieeexplore.ieee.org
… limited accuracy for detecting arbitrary shape text. In this … shape scenetextdetector through boundary pointsdynamic … a text aware module (TAM) and a boundary pointdynamic opti…
M Zhao, W Feng, F Yin, CL Liu - Pattern Recognition Letters, 2023 - Elsevier
… to combine center point annotation into mixed-supervised scenetextdetection, in which the … points. For better incorporating point supervision, we adopt self-training strategy based on a …
… surveyed scenetextdetection and recognition with advances in … achievements in the scene textrecognition field based on … potential fiducial points and tuning their positions dynamically, …
Y Chen, W Wang, Y Zhou, F Yang… - … Pattern Recognition …, 2021 - ieeexplore.ieee.org
… 5, the corresponding points of detection results are the inputs of linear interpolation function … multi-orientation scenetextdetection: A unified framework with dynamic programming,” TIP, …
… Figure 1: A unified integration framework with dynamic programming for scenetextdetection in video by learning locally and globally, where wV and wE are vertices and edges’ weights …
L Xiao, P Zhou, K Xu, X Zhao - Sensors, 2021 - mdpi.com
… points of each prediction frame need to be sorted and compared to obtain the first coordinate point (usually put the upper left corner point … real-time video detection and dynamic target …