Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -22,8 +22,138 @@ dataset_info:
|
|
22 |
dtype: int64
|
23 |
splits:
|
24 |
- name: train
|
25 |
-
num_bytes: 16755833083
|
26 |
num_examples: 14394510
|
27 |
download_size: 5410262648
|
28 |
-
dataset_size: 16755833083
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
29 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
dtype: int64
|
23 |
splits:
|
24 |
- name: train
|
25 |
+
num_bytes: 16755833083
|
26 |
num_examples: 14394510
|
27 |
download_size: 5410262648
|
28 |
+
dataset_size: 16755833083
|
29 |
+
task_categories:
|
30 |
+
- text-to-video
|
31 |
+
- video-classification
|
32 |
+
language:
|
33 |
+
- en
|
34 |
+
tags:
|
35 |
+
- captions
|
36 |
+
- metadata
|
37 |
+
pretty_name: CleanVid Map (15M)
|
38 |
+
size_categories:
|
39 |
+
- 10M<n<100M
|
40 |
---
|
41 |
+
|
42 |
+
# CleanVid Map (15M) π₯
|
43 |
+
### TempoFunk Video Generation
|
44 |
+
|
45 |
+
CleanVid-15M is a large-scale dataset of videos with multiple metadata entries such as:
|
46 |
+
- Textual Descriptions π
|
47 |
+
- Recording Equipment πΉ
|
48 |
+
- Categories π
|
49 |
+
- Framerate ποΈ
|
50 |
+
- Aspect Ratio πΊ
|
51 |
+
|
52 |
+
CleanVid aim is to improve the quality of WebVid-10M dataset by adding more data and cleaning the dataset by dewatermarking the videos in it.
|
53 |
+
|
54 |
+
This dataset includes only the map with the urls and metadata, with 3,694,510 more entries than the original WebVid-10M dataset.
|
55 |
+
|
56 |
+
CleanVid is the foundation dataset for the TempoFunk Video Generation project.
|
57 |
+
|
58 |
+
CleanVid is built from a crawl of Shutterstock from June 25, 2023.
|
59 |
+
|
60 |
+
## Format π
|
61 |
+
|
62 |
+
- id: Integer (int64) - Shutterstock video ID
|
63 |
+
- description: String - Description of the video
|
64 |
+
- duration: Float(64) - Duration of the video in seconds
|
65 |
+
- aspectratio: String - Aspect Ratio of the video separated by colons (":")
|
66 |
+
- videourl: String - Video URL for the video in the entry, MP4 format. WEBM format is also available most of the times (by changing the extension at the end of the URL.).
|
67 |
+
- author: String - JSON-String containing information of the author such as `Recording Equipment`, `Style`, `Nationality` and others.
|
68 |
+
- categories: String - JSON-String containing the categories of the videos. (Values from shutterstock, not by us.)
|
69 |
+
- framerate: Float(64) - Framerate of the video
|
70 |
+
- r18: Bit (int64) - Wether the video is marked as mature content. 0 = Safe For Work; 1 = Mature Content
|
71 |
+
|
72 |
+
## Code π©βπ»
|
73 |
+
|
74 |
+
If you want to re-create this dataset on your own, code is available here:
|
75 |
+
|
76 |
+
https://github.com/chavinlo/tempofunk-scrapper/tree/refractor1/sites/shutterstock
|
77 |
+
|
78 |
+
Due to rate-limitations, you might need to obtain a proxy. Functionality for proxies is included in the repository.
|
79 |
+
|
80 |
+
## Sample π§ͺ
|
81 |
+
|
82 |
+
```json
|
83 |
+
{
|
84 |
+
"id": 1056934082,
|
85 |
+
"description": "Rio, Brazil - February 24, 2020: parade of the samba school Mangueira, at the Marques de Sapucai Sambodromo",
|
86 |
+
"duration": 9.76,
|
87 |
+
"aspectratio": "16:9",
|
88 |
+
"videourl": "https://www.shutterstock.com/shutterstock/videos/1056934082/preview/stock-footage-rio-brazil-february-parade-of-the-samba-school-mangueira-at-the-marques-de-sapucai.mp4",
|
89 |
+
"author": {
|
90 |
+
"accountsId": 101974372,
|
91 |
+
"contributorId": 62154,
|
92 |
+
"bio": "Sempre produzindo mais",
|
93 |
+
"location": "br",
|
94 |
+
"website": "www.dcpress.com.br",
|
95 |
+
"contributorTypeList": [
|
96 |
+
"photographer"
|
97 |
+
],
|
98 |
+
"equipmentList": [
|
99 |
+
"300mm f2.8",
|
100 |
+
"24-70mm",
|
101 |
+
"70-200mm",
|
102 |
+
"Nikon D7500 ",
|
103 |
+
"Nikon Df",
|
104 |
+
"Flashs Godox"
|
105 |
+
],
|
106 |
+
"styleList": [
|
107 |
+
"editorial",
|
108 |
+
"food",
|
109 |
+
"landscape"
|
110 |
+
],
|
111 |
+
"subjectMatterList": [
|
112 |
+
"photographer",
|
113 |
+
"people",
|
114 |
+
"nature",
|
115 |
+
"healthcare",
|
116 |
+
"food_and_drink"
|
117 |
+
],
|
118 |
+
"facebookUsername": "celso.pupo",
|
119 |
+
"googlePlusUsername": "celsopupo",
|
120 |
+
"twitterUsername": "celsopupo",
|
121 |
+
"storageKey": "/contributors/62154/avatars/thumb.jpg",
|
122 |
+
"cdnThumbPath": "/contributors/62154/avatars/thumb.jpg",
|
123 |
+
"displayName": "Celso Pupo",
|
124 |
+
"vanityUrlUsername": "rodrigues",
|
125 |
+
"portfolioUrlSuffix": "rodrigues",
|
126 |
+
"portfolioUrl": "https://www.shutterstock.com/g/rodrigues",
|
127 |
+
"instagramUsername": "celsopupo",
|
128 |
+
"hasPublicSets": true,
|
129 |
+
"instagramUrl": "https://www.instagram.com/celsopupo",
|
130 |
+
"facebookUrl": "https://www.facebook.com/celso.pupo",
|
131 |
+
"twitterUrl": "https://twitter.com/celsopupo"
|
132 |
+
},
|
133 |
+
"categories": [
|
134 |
+
"People"
|
135 |
+
],
|
136 |
+
"framerate": 29.97,
|
137 |
+
"r18": 0
|
138 |
+
}
|
139 |
+
```
|
140 |
+
|
141 |
+
## Credits π₯
|
142 |
+
|
143 |
+
### Main
|
144 |
+
|
145 |
+
- Lopho - Part of TempoFunk Video Generation
|
146 |
+
- Chavinlo - Part of TempoFunk Video Generation & CleanVid Crawling, Scraping and Formatting
|
147 |
+
|
148 |
+
```
|
149 |
+
@InProceedings{Bain21,
|
150 |
+
author = "Max Bain and Arsha Nagrani and G{\"u}l Varol and Andrew Zisserman",
|
151 |
+
title = "Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval",
|
152 |
+
booktitle = "IEEE International Conference on Computer Vision",
|
153 |
+
year = "2021",
|
154 |
+
}
|
155 |
+
```
|
156 |
+
|
157 |
+
### Extra
|
158 |
+
|
159 |
+
- Salt - Base Threading Code (2022)
|