T
Tong He
Researcher at University of Adelaide
Publications - 43
Citations - 8098
Tong He is an academic researcher from University of Adelaide. The author has contributed to research in topics: Computer science & Object detection. The author has an hindex of 17, co-authored 38 publications receiving 4292 citations. Previous affiliations of Tong He include Wuhan University & Chinese Academy of Sciences.
Papers
More filters
Proceedings ArticleDOI
FCOS: Fully Convolutional One-Stage Object Detection
TL;DR: For the first time, a much simpler and flexible detection framework achieving improved detection accuracy is demonstrated, and it is hoped that the proposed FCOS framework can serve as a simple and strong alternative for many other instance-level tasks.
Posted Content
FCOS: Fully Convolutional One-Stage Object Detection
TL;DR: In this paper, a fully convolutional one-stage object detector (FCOS) is proposed to solve object detection in a per-pixel prediction fashion, analogue to semantic segmentation.
Book ChapterDOI
Detecting Text in Natural Image with Connectionist Text Proposal Network
TL;DR: The Connectionist Text Proposal Network (CTPN) as mentioned in this paper detects a text line in a sequence of fine-scale text proposals directly in convolutional feature maps, and develops a vertical anchor mechanism that jointly predicts location and text/non-text score of each fixed-width proposal.
Posted Content
Detecting Text in Natural Image with Connectionist Text Proposal Network
TL;DR: A novel Connectionist Text Proposal Network (CTPN) that accurately localizes text lines in natural image and develops a vertical anchor mechanism that jointly predicts location and text/non-text score of each fixed-width proposal, considerably improving localization accuracy.
Journal ArticleDOI
Text-Attentional Convolutional Neural Network for Scene Text Detection
TL;DR: A new system for scene text detection by proposing a novel text-attentional convolutional neural network (Text-CNN) that particularly focuses on extracting text-related regions and features from the image components and a powerful low-level detector called contrast-enhancement maximally stable extremal regions (MSERs) is developed.