|
--- |
|
language: |
|
- vi |
|
- en |
|
pretty_name: COCO 2017 image captions in Vietnamese |
|
source-datasets: |
|
- ms coco |
|
tags: |
|
- coco |
|
- coco-2017-vi |
|
- Vietnamese |
|
license: unknown |
|
task_categories: |
|
- image-to-text |
|
task_ids: |
|
- image-captioning |
|
--- |
|
# COCO 2017 image captions in Vietnamese |
|
|
|
The dataset is firstly introduced in [dinhanhx/VisualRoBERTa](https://github.com/dinhanhx/VisualRoBERTa/tree/main). |
|
I use VinAI tools to translate [COCO 2027 image caption](https://cocodataset.org/#download) (2017 Train/Val annotations) from English to Vietnamese. |
|
Then we merge [UIT-ViIC](https://arxiv.org/abs/2002.00175) dataset into it. |
|
To load the dataset, one can take a look at [this code in VisualRoBERTa](https://github.com/dinhanhx/VisualRoBERTa/blob/main/src/data.py#L22-L100) or [this code in Velvet](https://github.com/dinhanhx/velvet/blob/main/velvet/dataset/coco.py). |
|
|
|
I provide both English original and Vietnamese version (including UIT-ViIC). |
|
|
|
⚠ Note: |
|
- UIT-ViIC splits are originated from `en/captions_train2017.json`. Therefore, I combine all UIT-ViIC splits then I merge into `vi/captions_train2017_trans.json`. As a result, I get `captions_train2017_trans_plus.json`. |
|
- `vi/captions_train2017_trans.json` and `vi/captions_val2017_trans.json` are VinAI-translated from the ones in `en/`. |
