Corpus ID: 235828717

Synthesis in Style: Semantic Segmentation of Historical Documents using Synthetic Data

@article{Bartz2021SynthesisIS,
  title={Synthesis in Style: Semantic Segmentation of Historical Documents using Synthetic Data},
  author={Christian Bartz and Hendrik R{\"a}tz and Haojin Yang and Joseph Bethge and Christoph Meinel},
  journal={ArXiv},
  year={2021},
  volume={abs/2107.06777}
}
One of the most pressing problems in the automated analysis of historical documents is the availability of annotated training data. In this paper, we propose a novel method for the synthesis of training data for semantic segmentation of document images. We utilize clusters found in intermediate features of a StyleGAN generator for the synthesis of RGB and label images at the same time. Our model can be applied to any dataset of scanned documents without the need for manual annotation of… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 33 REFERENCES
dhSegment: A Generic Deep-Learning Approach for Document Segmentation
TLDR
This paper proposes an open-source implementation of a CNN-based pixel-wise predictor coupled with task dependent post-processing blocks and shows that a single CNN-architecture can be used across tasks with competitive results. Expand
Document Image Page Segmentation and Character Recognition as Semantic Segmentation
TLDR
This work uses fully supervised Deep CNN semantic segmentation to separate content layers from historical document images containing diverse content types, including handwriting, machine print, form lines, and stamps, using CNNs for semantic pixel labeling. Expand
DIVA-HisDB: A Precisely Annotated Large Dataset of Challenging Medieval Manuscripts
TLDR
A publicly available historical manuscript database DIVA-HisDB is introduced for the evaluation of several Document Image Analysis (DIA) tasks and a layout analysis ground-truth which has been iterated on, reviewed, and refined by an expert in medieval studies is provided. Expand
Synthetic Data for the Analysis of Archival Documents: Handwriting Determination
TLDR
This paper presents an approach for determining whether a scan of a document contains handwriting, and introduces a data generation method to successfully train the proposed deep neural network. Expand
Reading Text in the Wild with Convolutional Neural Networks
TLDR
An end-to-end system for text spotting—localising and recognising text in natural scene images—and text based image retrieval and a real-world application to allow thousands of hours of news footage to be instantly searchable via a text query is demonstrated. Expand
READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents
TLDR
This paper collects and annotates 2036 archival document images from different locations and time periods and proposes a new evaluation scheme that is based on baselines, which has no need for binarization and it can handle skewed as well as rotated text lines. Expand
A Modular Region and Text Line Layout Analysis System
  • Benjamin Kiessling
  • Computer Science
  • 2020 17th International Conference on Frontiers in Handwriting Recognition (ICFHR)
  • 2020
TLDR
An artificial neural network based approach to prediction of either that is implemented as part of a libre optical character recognition package and highly reconfigurable for a variety of tasks is presented. Expand
HORAE: an annotated dataset of books of hours
TLDR
A new dataset of annotated pages from books of hours, a type of handwritten prayerbooks owned and used by rich lay people in the late middle ages, is introduced and the evaluation of a state-of-the-art system for text line detection and for zone detection and typing is presented. Expand
Editing in Style: Uncovering the Local Semantics of GANs
TLDR
A simple and effective method for making local, semantically-aware edits to a target output image via a novel manipulation of style vectors that relies on the emergent disentanglement of semantic objects learned by StyleGAN during its training. Expand
Start, Follow, Read: End-to-End Full-Page Handwriting Recognition
TLDR
A deep learning model that jointly learns text detection, segmentation, and recognition using mostly images without detection or segmentation annotations, which exceeds the performance of the winner of the ICDAR2017 handwriting recognition competition. Expand
...
1
2
3
4
...