• Publications
  • Influence
CBNet: A Novel Composite Backbone Network Architecture for Object Detection
TLDR
This paper proposes a novel strategy for assembling multiple identical backbones by composite connections between the adjacent backbones to form a more powerful backbone named Composite Backbone Network (CBNet), and shows that CBNet can be very easily integrated into most state-of-the-art detectors and significantly improve their performances. Expand
Dataset, Ground-Truth and Performance Metrics for Table Detection Evaluation
TLDR
A dataset that is representative, large and most importantly, publicly available, and the compatible format of the ground truth makes evaluation independent of document medium is provided. Expand
A mathematics retrieval system for formulae in layout presentations
TLDR
Experiments show that the new system along with novel algorithms, comparing with two representative mathematics retrieval systems, provides more efficient mathematical formula index and retrieval, while simplifying user query input for PDF documents. Expand
Mathematical Formula Identification in PDF Documents
TLDR
A novel method by combining rule-based and learning-based methods to detect both isolated and embedded mathematical expressions in PDF documents is proposed and successfully incorporated into a commercial software package for large-scale Chinese e-Book production. Expand
WikiMirs: a mathematical information retrieval system for wikipedia
TLDR
Experimental results show that WikiMirs can efficiently support sub-structure matching and similarity matching of mathematical formulae, and obtains both higher accuracy and better ranked results over Wikipedia in comparison to Wikipedia Search and Egomath. Expand
WikiMirs 3.0: A Hybrid MIR System Based on the Context, Structure and Importance of Formulae in a Document
TLDR
Experimental results, compared with two classical MIR systems, demonstrate that the proposed system along with the novel model provides higher accuracy and better ranking results over Wikipedia. Expand
An End-to-End Quadrilateral Regression Network for Comic Panel Extraction
TLDR
This work proposes an end-to-end, two-stage quadrilateral regressing network architecture for comic panel detection, which inherits the architecture of Faster R-CNN and demonstrates that the proposed method significantly outperforms the existing Comic panel detection methods on multiple datasets by F1-score and page accuracy. Expand
A Table Detection Method for Multipage PDF Documents via Visual Seperators and Tabular Structures
TLDR
A novel and effective table detection method via visual separators and geometric content layout information, targeting at PDF documents is proposed, successfully incorporated into a commercial software package for large-scale Chinese e-Book production. Expand
A Deep Learning-Based Formula Detection Method for PDF Documents
TLDR
The Convolutional Neural Network and Recurrent Neural Network model is combined and refined to detect formulae according to both their character and vision features and outperforms the existing formula detection methods on both a ground-truth dataset and a larger self-built dataset. Expand
Mutual Enhancement for Detection of Multiple Logos in Sports Videos
TLDR
This paper proposes a mutual-enhanced approach that can improve the detection of a logo through the information obtained from other simultaneously occurred logos, in a Fast-RCNN-based framework. Expand
...
1
2
3
4
5
...