zhenjiemao commited on
Commit
bd263b0
·
verified ·
1 Parent(s): 66ec39b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -4
README.md CHANGED
@@ -37,9 +37,12 @@ dataset_info:
37
  ---
38
 
39
  # Dataset Card for aRefCOCO
40
- [![Paper](https://img.shields.io/badge/Paper-NeurIPS%202025-red)](https://arxiv.org/pdf/2510.10160)
41
- [![GitHub](https://img.shields.io/badge/GitHub-Repository-black?logo=github)](https://github.com/zhenjiemao/SaFiRe)
42
- [![Hugging Face Datasets](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Datasets-blue)](https://huggingface.co/datasets/zhenjiemao/aRefCOCO)
 
 
 
43
  [![License: CC BY 4.0](https://img.shields.io/badge/License-CC--BY--4.0-lightgrey.svg)](https://creativecommons.org/licenses/by/4.0/)
44
 
45
  **aRefCOCO** (Ambiguous RefCOCO) is a dataset specifically constructed for Referring Image Segmentation (RIS), focusing on **referential ambiguity** that frequently arises in real-world application. It introduces **object-distracting expressions**, which involve multiple entities with contextual cues, and **category-implicit expressions**, where the object class is not explicitly stated. Each entity is paired with an image, a target segmentation mask, multiple referring descriptions, and supporting metadata such as bounding boxes and category labels. In addition to the original benchmark used for evaluation, aRefCOCO now provides an extended **train split** to support model training and further research on referential ambiguity in referring segmentation and related tasks.
@@ -99,7 +102,7 @@ print(f"Descriptions: {sample['descriptions']}")
99
 
100
  This Hugging Face repository contains the dataset in **Parquet/Arrow format** for easy loading.
101
 
102
- For alternative formats and implementations, please visit the **[GitHub Repository](https://github.com/zhenjiemao/SaFiRe)** which includes:
103
  - Custom PyTorch Dataset class (`refdataset/refdataset.py`)
104
  - Source images and masks in original quality
105
  - JSONL metadata files
 
37
  ---
38
 
39
  # Dataset Card for aRefCOCO
40
+ [![Project Page](https://img.shields.io/badge/Project-Page-green?logo=githubpages)](https://zhenjiemao.github.io/SaFiRe/)
41
+ [![Paper](https://img.shields.io/badge/Paper-NeurIPS%202025-red?)](https://arxiv.org/pdf/2510.10160)
42
+ [![arXiv](https://img.shields.io/badge/arXiv-paper-red?logo=arxiv)](https://arxiv.org/abs/2510.10160)
43
+ [![SaFiRe Model](https://img.shields.io/badge/Model-SaFiRe-black?logo=github)](https://github.com/zhenjiemao/SaFiRe)
44
+ [![aRefCOCO Dataset](https://img.shields.io/badge/Dataset-aRefCOCO-blue?logo=github)](https://github.com/zhenjiemao/aRefCOCO)
45
+ [![aRefCOCO Dataset](https://img.shields.io/badge/Dataset-aRefCOCO-yellow?logo=huggingface)](https://huggingface.co/datasets/zhenjiemao/aRefCOCO)
46
  [![License: CC BY 4.0](https://img.shields.io/badge/License-CC--BY--4.0-lightgrey.svg)](https://creativecommons.org/licenses/by/4.0/)
47
 
48
  **aRefCOCO** (Ambiguous RefCOCO) is a dataset specifically constructed for Referring Image Segmentation (RIS), focusing on **referential ambiguity** that frequently arises in real-world application. It introduces **object-distracting expressions**, which involve multiple entities with contextual cues, and **category-implicit expressions**, where the object class is not explicitly stated. Each entity is paired with an image, a target segmentation mask, multiple referring descriptions, and supporting metadata such as bounding boxes and category labels. In addition to the original benchmark used for evaluation, aRefCOCO now provides an extended **train split** to support model training and further research on referential ambiguity in referring segmentation and related tasks.
 
102
 
103
  This Hugging Face repository contains the dataset in **Parquet/Arrow format** for easy loading.
104
 
105
+ For alternative formats and implementations, please visit the **[GitHub Repository](https://github.com/zhenjiemao/aRefCOCO)** which includes:
106
  - Custom PyTorch Dataset class (`refdataset/refdataset.py`)
107
  - Source images and masks in original quality
108
  - JSONL metadata files