# UNITS: Unsupervised Intermediate Training Stage for Scene Text Detection

arXiv: https://arxiv.org/abs/2205.04683 · HTML: https://mlsanity.com/paper/2205.04683/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 32 references that are in the map.
- First posted on arXiv 10 May 2022.

## Builds on (most cited first, 27 of 27)

- [Unsupervised Data Augmentation for Consistency Training](https://mlsanity.com/paper/1904.12848/) (2019)
- [Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision](https://mlsanity.com/paper/2106.01226/) (2021)
- [Synthetic Data for Text Localisation in Natural Images](https://mlsanity.com/paper/1604.06646/) (2016)
- [Character Region Awareness for Text Detection](https://mlsanity.com/paper/1904.01941/) (2019)
- [EAST: An Efficient and Accurate Scene Text Detector](https://mlsanity.com/paper/1704.03155/) (2017)
- [Real-time Scene Text Detection with Differentiable Binarization](https://mlsanity.com/paper/1911.08947/) (2019)
- [PseudoSeg: Designing Pseudo Labels for Semantic Segmentation](https://mlsanity.com/paper/2010.09713/) (2020)
- [Arbitrary-Oriented Scene Text Detection via Rotation Proposals](https://mlsanity.com/paper/1703.01086/) (2017)
- [Shape Robust Text Detection with Progressive Scale Expansion Network](https://mlsanity.com/paper/1903.12473/) (2019)
- [SEED: Semantics Enhanced Encoder-Decoder Framework for Scene Text Recognition](https://mlsanity.com/paper/2005.10977/) (2020)
- [TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes](https://mlsanity.com/paper/1807.01544/) (2018)
- [Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network](https://mlsanity.com/paper/1908.05900/) (2019)
- [Video Playback Rate Perception for Self-supervisedSpatio-Temporal Representation Learning](https://mlsanity.com/paper/2006.11476/) (2020)
- [Video Cloze Procedure for Self-Supervised Spatio-Temporal Learning](https://mlsanity.com/paper/2001.00294/) (2020)
- [Detecting Oriented Text in Natural Images by Linking Segments](https://mlsanity.com/paper/1703.06520/) (2017)
- [Deep Relational Reasoning Graph Network for Arbitrary Shape Text Detection](https://mlsanity.com/paper/2003.07493/) (2020)
- [PIMNet: A Parallel, Iterative and Mimicking Network for Scene Text Recognition](https://mlsanity.com/paper/2109.04145/) (2021)
- [GA-DAN: Geometry-Aware Domain Adaptation Network for Scene Text Detection and Recognition](https://mlsanity.com/paper/1907.09653/) (2019)
- [Dense Semantic Contrast for Self-Supervised Visual Representation Learning](https://mlsanity.com/paper/2109.07756/) (2021)
- [Mask is All You Need: Rethinking Mask R-CNN for Dense and Arbitrary-Shaped Scene Text Detection](https://mlsanity.com/paper/2109.03426/) (2021)
- [Synthetic-to-Real Unsupervised Domain Adaptation for Scene Text Detection in the Wild](https://mlsanity.com/paper/2009.01766/) (2020)
- [Curved Text Detection in Natural Scene Images with Semi- and Weakly-Supervised Learning](https://mlsanity.com/paper/1908.09990/) (2019)
- [Self-Training for Domain Adaptive Scene Text Detection](https://mlsanity.com/paper/2005.11487/) (2020)
- [Exploring Relations in Untrimmed Videos for Self-Supervised Learning](https://mlsanity.com/paper/2008.02711/) (2020)
- [FC2RN: A Fully Convolutional Corner Refinement Network for Accurate Multi-Oriented Scene Text Detection](https://mlsanity.com/paper/2007.05113/) (2020)
- [Two-Level Residual Distillation based Triple Network for Incremental Object Detection](https://mlsanity.com/paper/2007.13428/) (2020)
- [Which and Where to Focus: A Simple yet Accurate Framework for Arbitrary-Shaped Nearby Text Detection in Scene Images](https://mlsanity.com/paper/2109.03451/) (2021)

## Built on by (most cited first, 2 of 2)

- [Towards Robust Real-Time Scene Text Detection: From Semantic to Instance Representation Learning](https://mlsanity.com/paper/2308.07202/) (2023)
- [Towards Training-Free Scene Text Editing](https://mlsanity.com/paper/2603.24571/) (2026)

## Abstract (arXiv)

Recent scene text detection methods are almost based on deep learning and data-driven. Synthetic data is commonly adopted for pre-training due to expensive annotation cost. However, there are obvious domain discrepancies between synthetic data and real-world data. It may lead to sub-optimal performance to directly adopt the model initialized by synthetic data in the fine-tuning stage. In this paper, we propose a new training paradigm for scene text detection, which introduces an UNsupervised Intermediate Training Stage (UNITS) that builds a buffer path to real-world data and can alleviate the gap between the pre-training stage and fine-tuning stage. Three training strategies are further explored to perceive information from real-world data in an unsupervised way. With UNITS, scene text detectors are improved without introducing any parameters and computations during inference. Extensive experimental results show consistent performance improvements on three public datasets.