Datasets:
Languages:
English
Size:
1K<n<10K
ArXiv:
Tags:
iiw
imageinwords
image-descriptions
image-captions
detailed-descriptions
hyper-detailed-descriptions
License:
Update README.md
Browse files
README.md
CHANGED
@@ -1,170 +1 @@
|
|
1 |
-
|
2 |
-
annotations_creators:
|
3 |
-
- expert-generated
|
4 |
-
- crowdsourced
|
5 |
-
license: cc-by-4.0
|
6 |
-
task_categories:
|
7 |
-
- image-to-text
|
8 |
-
- text-to-image
|
9 |
-
- object-detection
|
10 |
-
language:
|
11 |
-
- en
|
12 |
-
size_categories:
|
13 |
-
- 1K<n<10K
|
14 |
-
tags:
|
15 |
-
- iiw
|
16 |
-
- imageinwords
|
17 |
-
- image-descriptions
|
18 |
-
- image-captions
|
19 |
-
- detailed-descriptions
|
20 |
-
- hyper-detailed-descriptions
|
21 |
-
- object-descriptions
|
22 |
-
- object-detection
|
23 |
-
- object-labels
|
24 |
-
pretty_name: ImageInWords
|
25 |
-
multilinguality:
|
26 |
-
- monolingual
|
27 |
-
---
|
28 |
-
|
29 |
-
<h2>ImageInWords: Unlocking Hyper-Detailed Image Descriptions</h2>
|
30 |
-
|
31 |
-
Please visit the [webpage](https://google.github.io/imageinwords) for all the information about the IIW project, data downloads, visualizations, and much more.
|
32 |
-
|
33 |
-
<img src="https://github.com/google/imageinwords/blob/main/static/images/Abstract/1_white_background.png?raw=true">
|
34 |
-
<img src="https://github.com/google/imageinwords/blob/main/static/images/Abstract/2_white_background.png?raw=true">
|
35 |
-
|
36 |
-
Please reach out to iiw-dataset@google.com for thoughts/feedback/questions/collaborations.
|
37 |
-
|
38 |
-
<h3>🤗Hugging Face🤗</h3>
|
39 |
-
|
40 |
-
<li><a href="https://huggingface.co/datasets/google/imageinwords">Dataset</a></li>
|
41 |
-
|
42 |
-
```python
|
43 |
-
from datasets import load_dataset
|
44 |
-
|
45 |
-
# `name` can be one of: IIW-400, DCI_Test, DOCCI_Test, CM_3600, LocNar_Eval
|
46 |
-
# refer: https://github.com/google/imageinwords/tree/main/datasets
|
47 |
-
dataset = load_dataset('google/imageinwords', token=None, name="IIW-400", trust_remote_code=True)
|
48 |
-
```
|
49 |
-
|
50 |
-
<li><a href="https://huggingface.co/spaces/google/imageinwords-explorer">Dataset-Explorer</a></li>
|
51 |
-
|
52 |
-
|
53 |
-
## Dataset Description
|
54 |
-
|
55 |
-
- **Paper:** [arXiv](https://arxiv.org/abs/2405.02793)
|
56 |
-
- **Homepage:** https://google.github.io/imageinwords/
|
57 |
-
- **Point of Contact:** iiw-dataset@google.com
|
58 |
-
- **Dataset Explorer:** [ImageInWords-Explorer](https://huggingface.co/spaces/google/imageinwords-explorer)
|
59 |
-
|
60 |
-
### Dataset Summary
|
61 |
-
|
62 |
-
ImageInWords (IIW), a carefully designed human-in-the-loop annotation framework for curating hyper-detailed image descriptions and a new dataset resulting from this process.
|
63 |
-
We validate the framework through evaluations focused on the quality of the dataset and its utility for fine-tuning with considerations for readability, comprehensiveness, specificity, hallucinations, and human-likeness.
|
64 |
-
|
65 |
-
This Data Card describes a mixture of human annotated and machine generated data intended to help create and capture rich, hyper-detailed image descriptions.
|
66 |
-
|
67 |
-
IIW dataset has two parts: human annotations and model outputs. The main purposes of this dataset are:
|
68 |
-
1) to provide samples from SoTA human authored outputs to promote discussion on annotation guidelines to further improve the quality
|
69 |
-
2) to provide human SxS results and model outputs to promote development of automatic metrics to mimic human SxS judgements.
|
70 |
-
|
71 |
-
### Supported Tasks
|
72 |
-
|
73 |
-
Text-to-Image, Image-to-Text, Object Detection
|
74 |
-
|
75 |
-
### Languages
|
76 |
-
|
77 |
-
English
|
78 |
-
|
79 |
-
## Dataset Structure
|
80 |
-
|
81 |
-
### Data Instances
|
82 |
-
|
83 |
-
### Data Fields
|
84 |
-
|
85 |
-
For details on the datasets and output keys, please refer to our [GitHub data](https://github.com/google/imageinwords/tree/main/datasets) page inside the individual folders.
|
86 |
-
|
87 |
-
IIW:
|
88 |
-
- `image/key`
|
89 |
-
- `image/url`
|
90 |
-
- `IIW`: Human generated image description
|
91 |
-
- `IIW-P5B`: Machine generated image description
|
92 |
-
- `iiw-human-sxs-gpt4v` and `iiw-human-sxs-iiw-p5b`: human SxS metrics
|
93 |
-
- metrics/Comprehensiveness
|
94 |
-
- metrics/Specificity
|
95 |
-
- metrics/Hallucination
|
96 |
-
- metrics/First few line(s) as tldr
|
97 |
-
- metrics/Human Like
|
98 |
-
|
99 |
-
DCI:
|
100 |
-
- `image`
|
101 |
-
- `image/url`
|
102 |
-
- `ex_id`
|
103 |
-
- `IIW`: Human authored image description
|
104 |
-
- `metrics/Comprehensiveness`
|
105 |
-
- `metrics/Specificity`
|
106 |
-
- `metrics/Hallucination`
|
107 |
-
- `metrics/First few line(s) as tldr`
|
108 |
-
- `metrics/Human Like`
|
109 |
-
|
110 |
-
DOCCI:
|
111 |
-
- `image`
|
112 |
-
- `image/thumbnail_url`
|
113 |
-
- `IIW`: Human generated image description
|
114 |
-
- `DOCCI`: Image description from DOCCI
|
115 |
-
- `metrics/Comprehensiveness`
|
116 |
-
- `metrics/Specificity`
|
117 |
-
- `metrics/Hallucination`
|
118 |
-
- `metrics/First few line(s) as tldr`
|
119 |
-
- `metrics/Human Like`
|
120 |
-
|
121 |
-
LocNar:
|
122 |
-
- `image/key`
|
123 |
-
- `image/url`
|
124 |
-
- `IIW-P5B`: Machine generated image description
|
125 |
-
|
126 |
-
CM3600:
|
127 |
-
- `image/key`
|
128 |
-
- `image/url`
|
129 |
-
- `IIW-P5B`: Machine generated image description
|
130 |
-
|
131 |
-
Please note that all fields are string.
|
132 |
-
|
133 |
-
### Data Splits
|
134 |
-
|
135 |
-
Dataset | Size
|
136 |
-
---| ---:
|
137 |
-
IIW-400 | 400
|
138 |
-
DCI_Test | 112
|
139 |
-
DOCCI_Test | 100
|
140 |
-
LocNar_Eval | 1000
|
141 |
-
CM_3600 | 1000
|
142 |
-
|
143 |
-
### Annotations
|
144 |
-
|
145 |
-
#### Annotation process
|
146 |
-
|
147 |
-
Some text descriptions were written by human annotators and some were generated by machine models.
|
148 |
-
The metrics are all from human SxS.
|
149 |
-
|
150 |
-
### Personal and Sensitive Information
|
151 |
-
|
152 |
-
The images that were used for the descriptions and the machine generated text descriptions are checked (by algorithmic methods and manual inspection) for S/PII, pornographic content, and violence and any we found may contain such information have been filtered.
|
153 |
-
We asked that human annotators use an objective and respectful language for the image descriptions.
|
154 |
-
|
155 |
-
### Licensing Information
|
156 |
-
|
157 |
-
CC BY 4.0
|
158 |
-
|
159 |
-
### Citation Information
|
160 |
-
|
161 |
-
```
|
162 |
-
@misc{garg2024imageinwords,
|
163 |
-
title={ImageInWords: Unlocking Hyper-Detailed Image Descriptions},
|
164 |
-
author={Roopal Garg and Andrea Burns and Burcu Karagol Ayan and Yonatan Bitton and Ceslee Montgomery and Yasumasa Onoe and Andrew Bunner and Ranjay Krishna and Jason Baldridge and Radu Soricut},
|
165 |
-
year={2024},
|
166 |
-
eprint={2405.02793},
|
167 |
-
archivePrefix={arXiv},
|
168 |
-
primaryClass={cs.CV}
|
169 |
-
}
|
170 |
-
```
|
|
|
1 |
+
یک عکس زیبا کنار ساحل```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|