# Natural Scene Text Editing Based on AI

arXiv: https://arxiv.org/abs/2111.15475 · HTML: https://mlsanity.com/paper/2111.15475/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 21 references that are in the map.
- First posted on arXiv 26 Nov 2021.

## Builds on (most cited first, 16 of 16)

- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Context Encoders: Feature Learning by Inpainting](https://mlsanity.com/paper/1604.07379/) (2016)
- [Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform](https://mlsanity.com/paper/1804.02815/) (2018)
- [EAST: An Efficient and Accurate Scene Text Detector](https://mlsanity.com/paper/1704.03155/) (2017)
- [Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis](https://mlsanity.com/paper/1601.04589/) (2016)
- [Multi-Content GAN for Few-Shot Font Style Transfer](https://mlsanity.com/paper/1712.00516/) (2017)
- [STEFANN: Scene Text Editor using Font Adaptive Neural Network](https://mlsanity.com/paper/1903.01192/) (2019)
- [Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation](https://mlsanity.com/paper/1802.08948/) (2018)
- [Auto-Encoder Guided GAN for Chinese Calligraphy Synthesis](https://mlsanity.com/paper/1706.08789/) (2017)
- [Edit Probability for Scene Text Recognition](https://mlsanity.com/paper/1805.03384/) (2018)
- [From A to Z: Supervised Transfer of Style and Content Using Deep Neural Network Generators](https://mlsanity.com/paper/1603.02003/) (2016)
- [Deep Learning-based Forgery Attack on Document Images](https://mlsanity.com/paper/2102.00653/) (2021)
- [Learning Implicit Glyph Shape Representation](https://mlsanity.com/paper/2106.08573/) (2021)
- [Learning to Read by Spelling: Towards Unsupervised Text Recognition](https://mlsanity.com/paper/1809.08675/) (2018)
- [Domain Generalization for Document Authentication against Practical Recapturing Attacks](https://mlsanity.com/paper/2101.01404/) (2021)
- [A Multitask Network for Localization and Recognition of Text in Images](https://mlsanity.com/paper/1906.09266/) (2019)

## Abstract (arXiv)

In a recorded situation, textual information is crucial for scene interpretation and decision making. The ability to edit text directly on images has a number of advantages, including error correction, text restoration, and image reusability. This research shows how to change image text at the letter and digits level. I devised a two-part letters-digits network (LDN) to encode and decode digital images, as well as learn and transfer the font style of the source characters to the target characters. This method allows you to update the uppercase letters, lowercase letters and digits in the picture.