MIT-10M: A Large Scale Parallel Corpus of Multilingual Image Translation
This addresses dataset limitations for researchers and practitioners in multilingual image translation, though it is incremental as it focuses on data collection rather than algorithmic innovation.
The authors tackled the problem of limited scale, diversity, and quality in image translation datasets by introducing MIT-10M, a large-scale parallel corpus with over 10M image-text pairs across 14 languages, which tripled model performance compared to baselines.
Image Translation (IT) holds immense potential across diverse domains, enabling the translation of textual content within images into various languages. However, existing datasets often suffer from limitations in scale, diversity, and quality, hindering the development and evaluation of IT models. To address this issue, we introduce MIT-10M, a large-scale parallel corpus of multilingual image translation with over 10M image-text pairs derived from real-world data, which has undergone extensive data cleaning and multilingual translation validation. It contains 840K images in three sizes, 28 categories, tasks with three levels of difficulty and 14 languages image-text pairs, which is a considerable improvement on existing datasets. We conduct extensive experiments to evaluate and train models on MIT-10M. The experimental results clearly indicate that our dataset has higher adaptability when it comes to evaluating the performance of the models in tackling challenging and complex image translation tasks in the real world. Moreover, the performance of the model fine-tuned with MIT-10M has tripled compared to the baseline model, further confirming its superiority.