KaraKaraWitch commited on
Commit
8c6ad64
1 Parent(s): 472b3e1

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +116 -0
README.md CHANGED
@@ -1,3 +1,119 @@
1
  ---
2
  license: cc
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: cc
3
+ tags:
4
+ - art
5
+ - anime
6
+ pretty_name: LFAnime
7
+ task_categories:
8
+ - image-classification
9
+ - text-to-image
10
  ---
11
+
12
+ # Dataset Card for LFANIME
13
+
14
+ A dataset of anime frames collected by KaraKaraWitch.
15
+
16
+ ## Dataset Details
17
+
18
+ ### Dataset Description
19
+
20
+ LFANIME, or Low-Framerate Anime, comprises frames from Japanese animation. The dataset serves dual purposes—facilitating fine-tuning of image diffusion models and functioning as a pre-training resource. Moreover, we anticipate its utilization in image classification.
21
+
22
+ Important Note: LFAnime is not intended for watching anime. To discourage this application, we have intentionally lowered the frame rate and excluded audio from the dataset.
23
+
24
+ - **Curated by:** KaraKaraWitch
25
+ - **Funded by [optional]:** N/A
26
+ - **Shared by [optional]:** N/A
27
+ - **Language(s) (NLP):** Nil. Primarily japanese, but no audio is included.
28
+ - **License:** CC
29
+
30
+ ## Uses
31
+
32
+ A tar file compresses each "Episode," encompassing sequential anime frames. The dataset also incorporates chapters for episodes that have them. It's important to note that certain frame numbers may be absent intentionally.
33
+
34
+ ### Direct Use
35
+
36
+ <!-- This section describes suitable use cases for the dataset. -->
37
+
38
+ We release this dataset for free in the hopes that it could be used for text to image generation and/or image classification.
39
+
40
+ ### Out-of-Scope Use
41
+
42
+ Technically speaking, this dataset could be used to watch anime. However we do not recommend as such.
43
+ Additionally there could be unforseen usage that the author does not intend.
44
+
45
+ <!-- This section addresses misuse, malicious use, and uses that the dataset will not work well for. -->
46
+
47
+ ## Dataset Structure
48
+
49
+ <!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
50
+
51
+ [More Information Needed]
52
+
53
+ ## Dataset Creation
54
+
55
+ ### Curation Rationale
56
+
57
+ <!-- Motivation for the creation of this dataset. -->
58
+
59
+ The emphasis has been on developing models for generating images from text, particularly in the realm of creating "anime"-style visuals.
60
+ Examples of such models include Waifu Diffusion and NovelAI's SD 1.x models. Regrettably, these models tend to converge, resulting in a consistent aesthetic.
61
+ While this aesthetic may appeal to many users, it poses a challenge when attempting to diverge from or fine-tune the ingrained visual style of most SD 1.x models.
62
+
63
+ ### Source Data
64
+
65
+ <!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). -->
66
+
67
+ #### Data Collection and Processing
68
+
69
+ <!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
70
+
71
+ We've opted not to reveal the specific origins of the anime to establish a level of separation between the producers and this dataset.
72
+ Nevertheless, we can outline the processing steps as follows:
73
+
74
+ 1. Extract frames from the mkv file, sampling every 10 frames per second.
75
+ 2. Utilize scxvid to generate a timecode for identifying scene cuts.
76
+ 3. Exclude frames that precede or follow a scene cut (considering potential inclusion of 1/2 frames at each scene cut).
77
+ 4. Save the processed frames to a tar file.
78
+
79
+ #### Who are the source data producers?
80
+
81
+ We have decided not to disclose the exact sources.
82
+
83
+ ## Bias, Risks, and Limitations
84
+
85
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
86
+
87
+ As this dataset is a personal collection from KaraKaraWitch, it will have tendencies to generally not "Shonen" anime and will have female protagonists in general.
88
+
89
+ ### Recommendations
90
+
91
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
92
+
93
+ Users should be made aware of the risks, biases and limitations of the dataset.
94
+
95
+ ## Citation [optional]
96
+
97
+ ```
98
+ @misc{lfanime,
99
+ title = {LFAnime: A Low Framerate anime dataset.},
100
+ author = {KaraKaraWitch},
101
+ year = {2023},
102
+ howpublished = {\url{https://huggingface.co/datasets/RyokoExtra/LFANIME}},
103
+ }
104
+ ```
105
+
106
+
107
+ ## Glossary [optional]
108
+
109
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the dataset or dataset card. -->
110
+
111
+ Anime:
112
+
113
+ > Anime (Japanese: アニメ, IPA: [aꜜɲime]) is hand-drawn and computer-generated animation originating from Japan. Outside Japan and in English, anime refers specifically to animation produced in Japan.[1] However, in Japan and in Japanese, anime (a term derived from a shortening of the English word animation) describes all animated works, regardless of style or origin. Many works of animation with a similar style to Japanese animation are also produced outside Japan. Video games sometimes also feature themes and artstyles that can be considered as "anime".
114
+ > - Wikipedia
115
+
116
+ ### Contributions
117
+
118
+ - [@KaraKaraWitch (Twitter)](https://twitter.com/KaraKaraWitch) for gathering this dataset.
119
+ - [ChatGPT](https://chat.openai.com) rewording sentences in this datacard.