Papers by Sivaji Bandyopadhyay

2 papers
WAT2019: English-Hindi Translation on Hindi Visual Genome Dataset (D19-52)

Copied to clipboard

Challenge: A multimodal translation is a task of translating a source language to a target language . a parallel text corpus and images are used to represent the contextual details of the text .
Approach: They compare a multimodal approach to a parallel text corpus and image caption generation approach to translate text in English to Hindi as a part of WAT2019 shared task.
Outcome: The proposed approach improves the translation of English to Hindi in three tasks . the proposed system is based on a neural network and can be used in healthcare, government, disaster management, etc.
English to Hindi Multi-modal Neural Machine Translation and Hindi Image Captioning (D19-52)

Copied to clipboard

Challenge: Multi-modal translation is an emerging task of the MT community, where visual features of image combine with textual features of parallel source-target text to translate sentences.
Approach: They propose to use convolutional neural net-works and visual geometry to extract image features and attention-based Neural MachineTranslation (NMT) system for translation.
Outcome: The proposed multi-modal translation system improves translation quality and improves the quality of the captions of the images.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations