# Modulating and attending the source image during encoding improves Multimodal Translation

arXiv: https://arxiv.org/abs/1712.03449 · HTML: https://mlsanity.com/paper/1712.03449/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Dec 2017.

## Built on by (most cited first, 5 of 5)

- [A Survey on Multi-modal Machine Translation: Tasks, Methods and Challenges](https://mlsanity.com/paper/2405.12669/) (2024)
- [Deep soccer captioning with transformer: dataset, semantics-related losses, and multi-level evaluation](https://mlsanity.com/paper/2202.05728/) (2022)
- [Physics-assisted Generative Adversarial Network for X-Ray Tomography](https://mlsanity.com/paper/2204.03703/) (2022)
- [IITP at WAT 2021: System description for English-Hindi Multimodal Translation Task](https://mlsanity.com/paper/2107.01656/) (2021)
- [Supervised Visual Attention for Simultaneous Multimodal Machine Translation](https://mlsanity.com/paper/2201.09324/) (2022)

## Abstract (arXiv)

We propose a new and fully end-to-end approach for multimodal translation where the source text encoder modulates the entire visual input processing using conditional batch normalization, in order to compute the most informative image features for our task. Additionally, we propose a new attention mechanism derived from this original idea, where the attention model for the visual input is conditioned on the source text encoder representations. In the paper, we detail our models as well as the image analysis pipeline. Finally, we report experimental results. They are, as far as we know, the new state of the art on three different test sets.