# Scene Text Synthesis for Efficient and Effective Deep Network Training

arXiv: https://arxiv.org/abs/1901.09193 · HTML: https://mlsanity.com/paper/1901.09193/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Jan 2019.

## Built on by (most cited first, 14 of 14)

- [Unbalanced Feature Transport for Exemplar-based Image Translation](https://mlsanity.com/paper/2106.10482/) (2021)
- [Blind Image Super-Resolution via Contrastive Representation Learning](https://mlsanity.com/paper/2107.00708/) (2021)
- [GMLight: Lighting Estimation via Geometric Distribution Approximation](https://mlsanity.com/paper/2102.10244/) (2021)
- [Bi-level Feature Alignment for Versatile Image Translation and Manipulation](https://mlsanity.com/paper/2107.03021/) (2021)
- [Visual Text Processing: A Comprehensive Review and Unified Evaluation](https://mlsanity.com/paper/2504.21682/) (2025)
- [Sparse Needlets for Lighting Estimation with Spherical Transport Loss](https://mlsanity.com/paper/2106.13090/) (2021)
- [STRIVE: Scene Text Replacement In Videos](https://mlsanity.com/paper/2109.02762/) (2021)
- [Deep Monocular 3D Human Pose Estimation via Cascaded Dimension-Lifting](https://mlsanity.com/paper/2104.03520/) (2021)
- [Visual Text Meets Low-level Vision: A Comprehensive Survey on Visual Text Processing](https://mlsanity.com/paper/2402.03082/) (2024)
- [FBC-GAN: Diverse and Flexible Image Synthesis via Foreground-Background Composition](https://mlsanity.com/paper/2107.03166/) (2021)
- [Weakly Supervised Scene Text Generation for Low-resource Languages](https://mlsanity.com/paper/2306.14269/) (2023)
- [A Survey on Adversarial Image Synthesis](https://mlsanity.com/paper/2106.16056/) (2021)
- [Scene Text recognition with Full Normalization](https://mlsanity.com/paper/2109.01034/) (2021)
- [Enhancing Scene Text Detectors with Realistic Text Image Synthesis Using Diffusion Models](https://mlsanity.com/paper/2311.16555/) (2023)

## Abstract (arXiv)

A large amount of annotated training images is critical for training accurate and robust deep network models but the collection of a large amount of annotated training images is often time-consuming and costly. Image synthesis alleviates this constraint by generating annotated training images automatically by machines which has attracted increasing interest in the recent deep learning research. We develop an innovative image synthesis technique that composes annotated training images by realistically embedding foreground objects of interest (OOI) into background images. The proposed technique consists of two key components that in principle boost the usefulness of the synthesized images in deep network training. The first is context-aware semantic coherence which ensures that the OOI are placed around semantically coherent regions within the background image. The second is harmonious appearance adaptation which ensures that the embedded OOI are agreeable to the surrounding background from both geometry alignment and appearance realism. The proposed technique has been evaluated over two related but very different computer vision challenges, namely, scene text detection and scene text recognition. Experiments over a number of public datasets demonstrate the effectiveness of our proposed image synthesis technique - the use of our synthesized images in deep network training is capable of achieving similar or even better scene text detection and scene text recognition performance as compared with using real images.