@inproceedings{zhu-etal-2023-peit,
title = "{PEIT}: Bridging the Modality Gap with Pre-trained Models for End-to-End Image Translation",
author = "Zhu, Shaolin and
Li, Shangjie and
Lei, Yikun and
Xiong, Deyi",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.751/",
doi = "10.18653/v1/2023.acl-long.751",
pages = "13433--13447",
abstract = "Image translation is a task that translates an image containing text in the source language to the target language. One major challenge with image translation is the modality gap between visual text inputs and textual inputs/outputs of machine translation (MT). In this paper, we propose PEIT, an end-to-end image translation framework that bridges the modality gap with pre-trained models. It is composed of four essential components: a visual encoder, a shared encoder-decoder backbone network, a vision-text representation aligner equipped with the shared encoder and a cross-modal regularizer stacked over the shared decoder. Both the aligner and regularizer aim at reducing the modality gap. To train PEIT, we employ a two-stage pre-training strategy with an auxiliary MT task: (1) pre-training the MT model on the MT training data to initialize the shared encoder-decoder backbone network; and (2) pre-training PEIT with the aligner and regularizer on a synthesized dataset with rendered images containing text from the MT training data. In order to facilitate the evaluation of PEIT and promote research on image translation, we create a large-scale image translation corpus ECOIT containing 480K image-translation pairs via crowd-sourcing and manual post-editing from real-world images in the e-commerce domain. Experiments on the curated ECOIT benchmark dataset demonstrate that PEIT substantially outperforms both cascaded image translation systems (OCR+MT) and previous strong end-to-end image translation model, with fewer parameters and faster decoding speed."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhu-etal-2023-peit">
<titleInfo>
<title>PEIT: Bridging the Modality Gap with Pre-trained Models for End-to-End Image Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shaolin</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shangjie</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yikun</namePart>
<namePart type="family">Lei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Deyi</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Image translation is a task that translates an image containing text in the source language to the target language. One major challenge with image translation is the modality gap between visual text inputs and textual inputs/outputs of machine translation (MT). In this paper, we propose PEIT, an end-to-end image translation framework that bridges the modality gap with pre-trained models. It is composed of four essential components: a visual encoder, a shared encoder-decoder backbone network, a vision-text representation aligner equipped with the shared encoder and a cross-modal regularizer stacked over the shared decoder. Both the aligner and regularizer aim at reducing the modality gap. To train PEIT, we employ a two-stage pre-training strategy with an auxiliary MT task: (1) pre-training the MT model on the MT training data to initialize the shared encoder-decoder backbone network; and (2) pre-training PEIT with the aligner and regularizer on a synthesized dataset with rendered images containing text from the MT training data. In order to facilitate the evaluation of PEIT and promote research on image translation, we create a large-scale image translation corpus ECOIT containing 480K image-translation pairs via crowd-sourcing and manual post-editing from real-world images in the e-commerce domain. Experiments on the curated ECOIT benchmark dataset demonstrate that PEIT substantially outperforms both cascaded image translation systems (OCR+MT) and previous strong end-to-end image translation model, with fewer parameters and faster decoding speed.</abstract>
<identifier type="citekey">zhu-etal-2023-peit</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.751</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.751/</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>13433</start>
<end>13447</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PEIT: Bridging the Modality Gap with Pre-trained Models for End-to-End Image Translation
%A Zhu, Shaolin
%A Li, Shangjie
%A Lei, Yikun
%A Xiong, Deyi
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F zhu-etal-2023-peit
%X Image translation is a task that translates an image containing text in the source language to the target language. One major challenge with image translation is the modality gap between visual text inputs and textual inputs/outputs of machine translation (MT). In this paper, we propose PEIT, an end-to-end image translation framework that bridges the modality gap with pre-trained models. It is composed of four essential components: a visual encoder, a shared encoder-decoder backbone network, a vision-text representation aligner equipped with the shared encoder and a cross-modal regularizer stacked over the shared decoder. Both the aligner and regularizer aim at reducing the modality gap. To train PEIT, we employ a two-stage pre-training strategy with an auxiliary MT task: (1) pre-training the MT model on the MT training data to initialize the shared encoder-decoder backbone network; and (2) pre-training PEIT with the aligner and regularizer on a synthesized dataset with rendered images containing text from the MT training data. In order to facilitate the evaluation of PEIT and promote research on image translation, we create a large-scale image translation corpus ECOIT containing 480K image-translation pairs via crowd-sourcing and manual post-editing from real-world images in the e-commerce domain. Experiments on the curated ECOIT benchmark dataset demonstrate that PEIT substantially outperforms both cascaded image translation systems (OCR+MT) and previous strong end-to-end image translation model, with fewer parameters and faster decoding speed.
%R 10.18653/v1/2023.acl-long.751
%U https://aclanthology.org/2023.acl-long.751/
%U https://doi.org/10.18653/v1/2023.acl-long.751
%P 13433-13447
Markdown (Informal)
[PEIT: Bridging the Modality Gap with Pre-trained Models for End-to-End Image Translation](https://aclanthology.org/2023.acl-long.751/) (Zhu et al., ACL 2023)
ACL