Hugging Face's logo

Datasets:
hexuan21
/
VideoScore_legacy

License:
Dataset card Data Studio Files
xet
Community
hexuan21 commited on
Commit
32d2ddd
·
verified ·
1 Parent(s): b1d2a8f

upload subset_20k/test_600 's json, videos and frames

Browse files
Files changed (1) hide show
  1. data/subset_20k/test_600/data_worsen_gen.json +1298 -0
data/subset_20k/test_600/data_worsen_gen.json ADDED
@@ -0,0 +1,1298 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "id": "b204331",
4
+ "images": [
5
+ "b204331_00.jpg",
6
+ "b204331_01.jpg",
7
+ "b204331_02.jpg",
8
+ "b204331_03.jpg",
9
+ "b204331_04.jpg",
10
+ "b204331_05.jpg",
11
+ "b204331_06.jpg",
12
+ "b204331_07.jpg",
13
+ "b204331_08.jpg",
14
+ "b204331_09.jpg",
15
+ "b204331_10.jpg",
16
+ "b204331_11.jpg",
17
+ "b204331_12.jpg",
18
+ "b204331_13.jpg",
19
+ "b204331_14.jpg",
20
+ "b204331_15.jpg"
21
+ ],
22
+ "conversations": [
23
+ {
24
+ "from": "human",
25
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is the cell donates an empty electron to an oxygen molecule ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
26
+ },
27
+ {
28
+ "from": "gpt",
29
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
30
+ }
31
+ ]
32
+ },
33
+ {
34
+ "id": "b604211",
35
+ "images": [
36
+ "b604211_00.jpg",
37
+ "b604211_01.jpg",
38
+ "b604211_02.jpg",
39
+ "b604211_03.jpg",
40
+ "b604211_04.jpg",
41
+ "b604211_05.jpg",
42
+ "b604211_06.jpg",
43
+ "b604211_07.jpg",
44
+ "b604211_08.jpg",
45
+ "b604211_09.jpg",
46
+ "b604211_10.jpg",
47
+ "b604211_11.jpg",
48
+ "b604211_12.jpg",
49
+ "b604211_13.jpg",
50
+ "b604211_14.jpg",
51
+ "b604211_15.jpg"
52
+ ],
53
+ "conversations": [
54
+ {
55
+ "from": "human",
56
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is Create a vignette for a course whose theme is the importance of women in the job market where the title is Plural Voices: Women ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
57
+ },
58
+ {
59
+ "from": "gpt",
60
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
61
+ }
62
+ ]
63
+ },
64
+ {
65
+ "id": "b205316",
66
+ "images": [
67
+ "b205316_00.jpg",
68
+ "b205316_01.jpg",
69
+ "b205316_02.jpg",
70
+ "b205316_03.jpg",
71
+ "b205316_04.jpg",
72
+ "b205316_05.jpg",
73
+ "b205316_06.jpg",
74
+ "b205316_07.jpg",
75
+ "b205316_08.jpg",
76
+ "b205316_09.jpg",
77
+ "b205316_10.jpg",
78
+ "b205316_11.jpg",
79
+ "b205316_12.jpg",
80
+ "b205316_13.jpg",
81
+ "b205316_14.jpg",
82
+ "b205316_15.jpg"
83
+ ],
84
+ "conversations": [
85
+ {
86
+ "from": "human",
87
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is absolute reality stunning photo of a woman in cashmere, loose fitting, casual, black sweater with long sleeves on crosswalk in New York City ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
88
+ },
89
+ {
90
+ "from": "gpt",
91
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 2\n overall score: 1\n"
92
+ }
93
+ ]
94
+ },
95
+ {
96
+ "id": "b802489",
97
+ "images": [
98
+ "b802489_00.jpg",
99
+ "b802489_01.jpg",
100
+ "b802489_02.jpg",
101
+ "b802489_03.jpg",
102
+ "b802489_04.jpg",
103
+ "b802489_05.jpg",
104
+ "b802489_06.jpg",
105
+ "b802489_07.jpg",
106
+ "b802489_08.jpg",
107
+ "b802489_09.jpg",
108
+ "b802489_10.jpg",
109
+ "b802489_11.jpg",
110
+ "b802489_12.jpg",
111
+ "b802489_13.jpg",
112
+ "b802489_14.jpg",
113
+ "b802489_15.jpg",
114
+ "b802489_16.jpg",
115
+ "b802489_17.jpg",
116
+ "b802489_18.jpg",
117
+ "b802489_19.jpg",
118
+ "b802489_20.jpg",
119
+ "b802489_21.jpg",
120
+ "b802489_22.jpg",
121
+ "b802489_23.jpg"
122
+ ],
123
+ "conversations": [
124
+ {
125
+ "from": "human",
126
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is warrior raising sword in the rain. ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
127
+ },
128
+ {
129
+ "from": "gpt",
130
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
131
+ }
132
+ ]
133
+ },
134
+ {
135
+ "id": "b802941",
136
+ "images": [
137
+ "b802941_00.jpg",
138
+ "b802941_01.jpg",
139
+ "b802941_02.jpg",
140
+ "b802941_03.jpg",
141
+ "b802941_04.jpg",
142
+ "b802941_05.jpg",
143
+ "b802941_06.jpg",
144
+ "b802941_07.jpg",
145
+ "b802941_08.jpg",
146
+ "b802941_09.jpg",
147
+ "b802941_10.jpg",
148
+ "b802941_11.jpg",
149
+ "b802941_12.jpg",
150
+ "b802941_13.jpg",
151
+ "b802941_14.jpg",
152
+ "b802941_15.jpg",
153
+ "b802941_16.jpg",
154
+ "b802941_17.jpg",
155
+ "b802941_18.jpg",
156
+ "b802941_19.jpg",
157
+ "b802941_20.jpg",
158
+ "b802941_21.jpg",
159
+ "b802941_22.jpg",
160
+ "b802941_23.jpg"
161
+ ],
162
+ "conversations": [
163
+ {
164
+ "from": "human",
165
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is claymation weird human midsommar mask covreing face, high detail, lots of detail, sunset, colourful, weird, creepy, surreal, macro detail ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
166
+ },
167
+ {
168
+ "from": "gpt",
169
+ "value": "visual quality: 1\n object consistency: 1\n dynamic degree: 2\n motion smoothness: 3\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
170
+ }
171
+ ]
172
+ },
173
+ {
174
+ "id": "b602727",
175
+ "images": [
176
+ "b602727_00.jpg",
177
+ "b602727_01.jpg",
178
+ "b602727_02.jpg",
179
+ "b602727_03.jpg",
180
+ "b602727_04.jpg",
181
+ "b602727_05.jpg",
182
+ "b602727_06.jpg",
183
+ "b602727_07.jpg",
184
+ "b602727_08.jpg",
185
+ "b602727_09.jpg",
186
+ "b602727_10.jpg",
187
+ "b602727_11.jpg",
188
+ "b602727_12.jpg",
189
+ "b602727_13.jpg",
190
+ "b602727_14.jpg",
191
+ "b602727_15.jpg"
192
+ ],
193
+ "conversations": [
194
+ {
195
+ "from": "human",
196
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is drummer of tool plays on piano ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
197
+ },
198
+ {
199
+ "from": "gpt",
200
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
201
+ }
202
+ ]
203
+ },
204
+ {
205
+ "id": "b304507",
206
+ "images": [
207
+ "b304507_00.jpg",
208
+ "b304507_01.jpg",
209
+ "b304507_02.jpg",
210
+ "b304507_03.jpg",
211
+ "b304507_04.jpg",
212
+ "b304507_05.jpg",
213
+ "b304507_06.jpg",
214
+ "b304507_07.jpg",
215
+ "b304507_08.jpg",
216
+ "b304507_09.jpg",
217
+ "b304507_10.jpg",
218
+ "b304507_11.jpg",
219
+ "b304507_12.jpg",
220
+ "b304507_13.jpg",
221
+ "b304507_14.jpg",
222
+ "b304507_15.jpg"
223
+ ],
224
+ "conversations": [
225
+ {
226
+ "from": "human",
227
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is the parents are fighting, the child in the middle is crying ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
228
+ },
229
+ {
230
+ "from": "gpt",
231
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 1\n overall score: 1\n"
232
+ }
233
+ ]
234
+ },
235
+ {
236
+ "id": "b005210",
237
+ "images": [
238
+ "b005210_00.jpg",
239
+ "b005210_01.jpg",
240
+ "b005210_02.jpg",
241
+ "b005210_03.jpg",
242
+ "b005210_04.jpg",
243
+ "b005210_05.jpg",
244
+ "b005210_06.jpg",
245
+ "b005210_07.jpg",
246
+ "b005210_08.jpg",
247
+ "b005210_09.jpg",
248
+ "b005210_10.jpg",
249
+ "b005210_11.jpg",
250
+ "b005210_12.jpg",
251
+ "b005210_13.jpg",
252
+ "b005210_14.jpg",
253
+ "b005210_15.jpg",
254
+ "b005210_16.jpg",
255
+ "b005210_17.jpg",
256
+ "b005210_18.jpg",
257
+ "b005210_19.jpg",
258
+ "b005210_20.jpg",
259
+ "b005210_21.jpg",
260
+ "b005210_22.jpg",
261
+ "b005210_23.jpg"
262
+ ],
263
+ "conversations": [
264
+ {
265
+ "from": "human",
266
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is a golden king monkey, empire, cute colors, epic, cinematic ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
267
+ },
268
+ {
269
+ "from": "gpt",
270
+ "value": "visual quality: 2\n object consistency: 1\n dynamic degree: 3\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
271
+ }
272
+ ]
273
+ },
274
+ {
275
+ "id": "b303570",
276
+ "images": [
277
+ "b303570_00.jpg",
278
+ "b303570_01.jpg",
279
+ "b303570_02.jpg",
280
+ "b303570_03.jpg",
281
+ "b303570_04.jpg",
282
+ "b303570_05.jpg",
283
+ "b303570_06.jpg",
284
+ "b303570_07.jpg",
285
+ "b303570_08.jpg",
286
+ "b303570_09.jpg",
287
+ "b303570_10.jpg",
288
+ "b303570_11.jpg",
289
+ "b303570_12.jpg",
290
+ "b303570_13.jpg",
291
+ "b303570_14.jpg",
292
+ "b303570_15.jpg"
293
+ ],
294
+ "conversations": [
295
+ {
296
+ "from": "human",
297
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is The camera slowly moves away from the cross as the sun sets on the horizon ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
298
+ },
299
+ {
300
+ "from": "gpt",
301
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 1\n factual consistency: 1\n overall score: 1\n"
302
+ }
303
+ ]
304
+ },
305
+ {
306
+ "id": "b304106",
307
+ "images": [
308
+ "b304106_00.jpg",
309
+ "b304106_01.jpg",
310
+ "b304106_02.jpg",
311
+ "b304106_03.jpg",
312
+ "b304106_04.jpg",
313
+ "b304106_05.jpg",
314
+ "b304106_06.jpg",
315
+ "b304106_07.jpg",
316
+ "b304106_08.jpg",
317
+ "b304106_09.jpg",
318
+ "b304106_10.jpg",
319
+ "b304106_11.jpg",
320
+ "b304106_12.jpg",
321
+ "b304106_13.jpg",
322
+ "b304106_14.jpg",
323
+ "b304106_15.jpg"
324
+ ],
325
+ "conversations": [
326
+ {
327
+ "from": "human",
328
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is cinematic vibrant action scene, two men in a chase crash through the window of the building, inception, matrix looking character HD, 4k resolution, sharp, red dragon cinema camera ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
329
+ },
330
+ {
331
+ "from": "gpt",
332
+ "value": "visual quality: 3\n object consistency: 1\n dynamic degree: 3\n motion smoothness: 2\n text-to-video alignment: 1\n factual consistency: 2\n overall score: 1\n"
333
+ }
334
+ ]
335
+ },
336
+ {
337
+ "id": "b305597",
338
+ "images": [
339
+ "b305597_00.jpg",
340
+ "b305597_01.jpg",
341
+ "b305597_02.jpg",
342
+ "b305597_03.jpg",
343
+ "b305597_04.jpg",
344
+ "b305597_05.jpg",
345
+ "b305597_06.jpg",
346
+ "b305597_07.jpg",
347
+ "b305597_08.jpg",
348
+ "b305597_09.jpg",
349
+ "b305597_10.jpg",
350
+ "b305597_11.jpg",
351
+ "b305597_12.jpg",
352
+ "b305597_13.jpg",
353
+ "b305597_14.jpg",
354
+ "b305597_15.jpg"
355
+ ],
356
+ "conversations": [
357
+ {
358
+ "from": "human",
359
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is merry Christmas with all the classic characters of Christmas with a happy holidays sign ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
360
+ },
361
+ {
362
+ "from": "gpt",
363
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 1\n"
364
+ }
365
+ ]
366
+ },
367
+ {
368
+ "id": "b602503",
369
+ "images": [
370
+ "b602503_00.jpg",
371
+ "b602503_01.jpg",
372
+ "b602503_02.jpg",
373
+ "b602503_03.jpg",
374
+ "b602503_04.jpg",
375
+ "b602503_05.jpg",
376
+ "b602503_06.jpg",
377
+ "b602503_07.jpg",
378
+ "b602503_08.jpg",
379
+ "b602503_09.jpg",
380
+ "b602503_10.jpg",
381
+ "b602503_11.jpg",
382
+ "b602503_12.jpg",
383
+ "b602503_13.jpg",
384
+ "b602503_14.jpg",
385
+ "b602503_15.jpg"
386
+ ],
387
+ "conversations": [
388
+ {
389
+ "from": "human",
390
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is man seeing the stars at night time ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
391
+ },
392
+ {
393
+ "from": "gpt",
394
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
395
+ }
396
+ ]
397
+ },
398
+ {
399
+ "id": "b005524",
400
+ "images": [
401
+ "b005524_00.jpg",
402
+ "b005524_01.jpg",
403
+ "b005524_02.jpg",
404
+ "b005524_03.jpg",
405
+ "b005524_04.jpg",
406
+ "b005524_05.jpg",
407
+ "b005524_06.jpg",
408
+ "b005524_07.jpg",
409
+ "b005524_08.jpg",
410
+ "b005524_09.jpg",
411
+ "b005524_10.jpg",
412
+ "b005524_11.jpg",
413
+ "b005524_12.jpg",
414
+ "b005524_13.jpg",
415
+ "b005524_14.jpg",
416
+ "b005524_15.jpg",
417
+ "b005524_16.jpg",
418
+ "b005524_17.jpg",
419
+ "b005524_18.jpg",
420
+ "b005524_19.jpg",
421
+ "b005524_20.jpg",
422
+ "b005524_21.jpg",
423
+ "b005524_22.jpg",
424
+ "b005524_23.jpg"
425
+ ],
426
+ "conversations": [
427
+ {
428
+ "from": "human",
429
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is The camera returns to the Grand Gallery, where art lovers continue to wander in awe. ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
430
+ },
431
+ {
432
+ "from": "gpt",
433
+ "value": "visual quality: 3\n object consistency: 1\n dynamic degree: 2\n motion smoothness: 3\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
434
+ }
435
+ ]
436
+ },
437
+ {
438
+ "id": "b105704",
439
+ "images": [
440
+ "b105704_00.jpg",
441
+ "b105704_01.jpg",
442
+ "b105704_02.jpg",
443
+ "b105704_03.jpg",
444
+ "b105704_04.jpg",
445
+ "b105704_05.jpg",
446
+ "b105704_06.jpg",
447
+ "b105704_07.jpg",
448
+ "b105704_08.jpg",
449
+ "b105704_09.jpg",
450
+ "b105704_10.jpg",
451
+ "b105704_11.jpg",
452
+ "b105704_12.jpg",
453
+ "b105704_13.jpg",
454
+ "b105704_14.jpg",
455
+ "b105704_15.jpg"
456
+ ],
457
+ "conversations": [
458
+ {
459
+ "from": "human",
460
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is A sheep in the style of cyberpunk driving ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
461
+ },
462
+ {
463
+ "from": "gpt",
464
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
465
+ }
466
+ ]
467
+ },
468
+ {
469
+ "id": "b702417",
470
+ "images": [
471
+ "b702417_00.jpg",
472
+ "b702417_01.jpg",
473
+ "b702417_02.jpg",
474
+ "b702417_03.jpg",
475
+ "b702417_04.jpg",
476
+ "b702417_05.jpg",
477
+ "b702417_06.jpg",
478
+ "b702417_07.jpg"
479
+ ],
480
+ "conversations": [
481
+ {
482
+ "from": "human",
483
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is man walking into a fantasy world, cinematic lighting ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> \n"
484
+ },
485
+ {
486
+ "from": "gpt",
487
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 1\n"
488
+ }
489
+ ]
490
+ },
491
+ {
492
+ "id": "b105684",
493
+ "images": [
494
+ "b105684_00.jpg",
495
+ "b105684_01.jpg",
496
+ "b105684_02.jpg",
497
+ "b105684_03.jpg",
498
+ "b105684_04.jpg",
499
+ "b105684_05.jpg",
500
+ "b105684_06.jpg",
501
+ "b105684_07.jpg",
502
+ "b105684_08.jpg",
503
+ "b105684_09.jpg",
504
+ "b105684_10.jpg",
505
+ "b105684_11.jpg",
506
+ "b105684_12.jpg",
507
+ "b105684_13.jpg",
508
+ "b105684_14.jpg",
509
+ "b105684_15.jpg"
510
+ ],
511
+ "conversations": [
512
+ {
513
+ "from": "human",
514
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is A love hate dance thats hard to navigate ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
515
+ },
516
+ {
517
+ "from": "gpt",
518
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 2\n factual consistency: 2\n overall score: 1\n"
519
+ }
520
+ ]
521
+ },
522
+ {
523
+ "id": "b801197",
524
+ "images": [
525
+ "b801197_00.jpg",
526
+ "b801197_01.jpg",
527
+ "b801197_02.jpg",
528
+ "b801197_03.jpg",
529
+ "b801197_04.jpg",
530
+ "b801197_05.jpg",
531
+ "b801197_06.jpg",
532
+ "b801197_07.jpg",
533
+ "b801197_08.jpg",
534
+ "b801197_09.jpg",
535
+ "b801197_10.jpg",
536
+ "b801197_11.jpg",
537
+ "b801197_12.jpg",
538
+ "b801197_13.jpg",
539
+ "b801197_14.jpg",
540
+ "b801197_15.jpg",
541
+ "b801197_16.jpg",
542
+ "b801197_17.jpg",
543
+ "b801197_18.jpg",
544
+ "b801197_19.jpg",
545
+ "b801197_20.jpg",
546
+ "b801197_21.jpg",
547
+ "b801197_22.jpg",
548
+ "b801197_23.jpg"
549
+ ],
550
+ "conversations": [
551
+ {
552
+ "from": "human",
553
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is the man flying slowly and too clouds ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
554
+ },
555
+ {
556
+ "from": "gpt",
557
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
558
+ }
559
+ ]
560
+ },
561
+ {
562
+ "id": "b703310",
563
+ "images": [
564
+ "b703310_00.jpg",
565
+ "b703310_01.jpg",
566
+ "b703310_02.jpg",
567
+ "b703310_03.jpg",
568
+ "b703310_04.jpg",
569
+ "b703310_05.jpg",
570
+ "b703310_06.jpg",
571
+ "b703310_07.jpg"
572
+ ],
573
+ "conversations": [
574
+ {
575
+ "from": "human",
576
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is golden skeletons rapping into a fisheye camera ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> \n"
577
+ },
578
+ {
579
+ "from": "gpt",
580
+ "value": "visual quality: 1\n object consistency: 1\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 2\n overall score: 1\n"
581
+ }
582
+ ]
583
+ },
584
+ {
585
+ "id": "b305385",
586
+ "images": [
587
+ "b305385_00.jpg",
588
+ "b305385_01.jpg",
589
+ "b305385_02.jpg",
590
+ "b305385_03.jpg",
591
+ "b305385_04.jpg",
592
+ "b305385_05.jpg",
593
+ "b305385_06.jpg",
594
+ "b305385_07.jpg",
595
+ "b305385_08.jpg",
596
+ "b305385_09.jpg",
597
+ "b305385_10.jpg",
598
+ "b305385_11.jpg",
599
+ "b305385_12.jpg",
600
+ "b305385_13.jpg",
601
+ "b305385_14.jpg",
602
+ "b305385_15.jpg"
603
+ ],
604
+ "conversations": [
605
+ {
606
+ "from": "human",
607
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is a lonely ghost girl standing in dark ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
608
+ },
609
+ {
610
+ "from": "gpt",
611
+ "value": "visual quality: 1\n object consistency: 1\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
612
+ }
613
+ ]
614
+ },
615
+ {
616
+ "id": "b106512",
617
+ "images": [
618
+ "b106512_00.jpg",
619
+ "b106512_01.jpg",
620
+ "b106512_02.jpg",
621
+ "b106512_03.jpg",
622
+ "b106512_04.jpg",
623
+ "b106512_05.jpg",
624
+ "b106512_06.jpg",
625
+ "b106512_07.jpg",
626
+ "b106512_08.jpg",
627
+ "b106512_09.jpg",
628
+ "b106512_10.jpg",
629
+ "b106512_11.jpg",
630
+ "b106512_12.jpg",
631
+ "b106512_13.jpg",
632
+ "b106512_14.jpg",
633
+ "b106512_15.jpg"
634
+ ],
635
+ "conversations": [
636
+ {
637
+ "from": "human",
638
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is Real footage of waves rising on a beach that is geographically opposite to where the Moon is in the sky. ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
639
+ },
640
+ {
641
+ "from": "gpt",
642
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
643
+ }
644
+ ]
645
+ },
646
+ {
647
+ "id": "b106941",
648
+ "images": [
649
+ "b106941_00.jpg",
650
+ "b106941_01.jpg",
651
+ "b106941_02.jpg",
652
+ "b106941_03.jpg",
653
+ "b106941_04.jpg",
654
+ "b106941_05.jpg",
655
+ "b106941_06.jpg",
656
+ "b106941_07.jpg",
657
+ "b106941_08.jpg",
658
+ "b106941_09.jpg",
659
+ "b106941_10.jpg",
660
+ "b106941_11.jpg",
661
+ "b106941_12.jpg",
662
+ "b106941_13.jpg",
663
+ "b106941_14.jpg",
664
+ "b106941_15.jpg"
665
+ ],
666
+ "conversations": [
667
+ {
668
+ "from": "human",
669
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is a sant give the gift to the children ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
670
+ },
671
+ {
672
+ "from": "gpt",
673
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 3\n factual consistency: 2\n overall score: 1\n"
674
+ }
675
+ ]
676
+ },
677
+ {
678
+ "id": "b204082",
679
+ "images": [
680
+ "b204082_00.jpg",
681
+ "b204082_01.jpg",
682
+ "b204082_02.jpg",
683
+ "b204082_03.jpg",
684
+ "b204082_04.jpg",
685
+ "b204082_05.jpg",
686
+ "b204082_06.jpg",
687
+ "b204082_07.jpg",
688
+ "b204082_08.jpg",
689
+ "b204082_09.jpg",
690
+ "b204082_10.jpg",
691
+ "b204082_11.jpg",
692
+ "b204082_12.jpg",
693
+ "b204082_13.jpg",
694
+ "b204082_14.jpg",
695
+ "b204082_15.jpg"
696
+ ],
697
+ "conversations": [
698
+ {
699
+ "from": "human",
700
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is travel in time in new york ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
701
+ },
702
+ {
703
+ "from": "gpt",
704
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 2\n motion smoothness: 3\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
705
+ }
706
+ ]
707
+ },
708
+ {
709
+ "id": "b703602",
710
+ "images": [
711
+ "b703602_00.jpg",
712
+ "b703602_01.jpg",
713
+ "b703602_02.jpg",
714
+ "b703602_03.jpg",
715
+ "b703602_04.jpg",
716
+ "b703602_05.jpg",
717
+ "b703602_06.jpg",
718
+ "b703602_07.jpg"
719
+ ],
720
+ "conversations": [
721
+ {
722
+ "from": "human",
723
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is man sleeping with a pitbull on bed ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> \n"
724
+ },
725
+ {
726
+ "from": "gpt",
727
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 3\n factual consistency: 1\n overall score: 1\n"
728
+ }
729
+ ]
730
+ },
731
+ {
732
+ "id": "b304442",
733
+ "images": [
734
+ "b304442_00.jpg",
735
+ "b304442_01.jpg",
736
+ "b304442_02.jpg",
737
+ "b304442_03.jpg",
738
+ "b304442_04.jpg",
739
+ "b304442_05.jpg",
740
+ "b304442_06.jpg",
741
+ "b304442_07.jpg",
742
+ "b304442_08.jpg",
743
+ "b304442_09.jpg",
744
+ "b304442_10.jpg",
745
+ "b304442_11.jpg",
746
+ "b304442_12.jpg",
747
+ "b304442_13.jpg",
748
+ "b304442_14.jpg",
749
+ "b304442_15.jpg"
750
+ ],
751
+ "conversations": [
752
+ {
753
+ "from": "human",
754
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is INTERTET ANIMATION EXPANDING AROUND THE WORLD ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
755
+ },
756
+ {
757
+ "from": "gpt",
758
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
759
+ }
760
+ ]
761
+ },
762
+ {
763
+ "id": "b801125",
764
+ "images": [
765
+ "b801125_00.jpg",
766
+ "b801125_01.jpg",
767
+ "b801125_02.jpg",
768
+ "b801125_03.jpg",
769
+ "b801125_04.jpg",
770
+ "b801125_05.jpg",
771
+ "b801125_06.jpg",
772
+ "b801125_07.jpg",
773
+ "b801125_08.jpg",
774
+ "b801125_09.jpg",
775
+ "b801125_10.jpg",
776
+ "b801125_11.jpg",
777
+ "b801125_12.jpg",
778
+ "b801125_13.jpg",
779
+ "b801125_14.jpg",
780
+ "b801125_15.jpg",
781
+ "b801125_16.jpg",
782
+ "b801125_17.jpg",
783
+ "b801125_18.jpg",
784
+ "b801125_19.jpg",
785
+ "b801125_20.jpg",
786
+ "b801125_21.jpg",
787
+ "b801125_22.jpg",
788
+ "b801125_23.jpg"
789
+ ],
790
+ "conversations": [
791
+ {
792
+ "from": "human",
793
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is Two characters talking to each other, father and son, cartoon, 3D, full, HDR, very high resolution, no distortion in the character, enhancement, direct with artificial intelligence technology, no errors, wonderful colors, and everything is good. ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
794
+ },
795
+ {
796
+ "from": "gpt",
797
+ "value": "visual quality: 1\n object consistency: 1\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 2\n factual consistency: 1\n overall score: 1\n"
798
+ }
799
+ ]
800
+ },
801
+ {
802
+ "id": "b403030",
803
+ "images": [
804
+ "b403030_00.jpg",
805
+ "b403030_01.jpg",
806
+ "b403030_02.jpg",
807
+ "b403030_03.jpg",
808
+ "b403030_04.jpg",
809
+ "b403030_05.jpg",
810
+ "b403030_06.jpg",
811
+ "b403030_07.jpg",
812
+ "b403030_08.jpg",
813
+ "b403030_09.jpg",
814
+ "b403030_10.jpg",
815
+ "b403030_11.jpg",
816
+ "b403030_12.jpg",
817
+ "b403030_13.jpg",
818
+ "b403030_14.jpg",
819
+ "b403030_15.jpg"
820
+ ],
821
+ "conversations": [
822
+ {
823
+ "from": "human",
824
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is gold harp frame on grey light wall ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
825
+ },
826
+ {
827
+ "from": "gpt",
828
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 1\n"
829
+ }
830
+ ]
831
+ },
832
+ {
833
+ "id": "b204246",
834
+ "images": [
835
+ "b204246_00.jpg",
836
+ "b204246_01.jpg",
837
+ "b204246_02.jpg",
838
+ "b204246_03.jpg",
839
+ "b204246_04.jpg",
840
+ "b204246_05.jpg",
841
+ "b204246_06.jpg",
842
+ "b204246_07.jpg",
843
+ "b204246_08.jpg",
844
+ "b204246_09.jpg",
845
+ "b204246_10.jpg",
846
+ "b204246_11.jpg",
847
+ "b204246_12.jpg",
848
+ "b204246_13.jpg",
849
+ "b204246_14.jpg",
850
+ "b204246_15.jpg"
851
+ ],
852
+ "conversations": [
853
+ {
854
+ "from": "human",
855
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is The bearded merman sitting on rocks in the ocean,Highest quality, very clear, a masterpiece ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
856
+ },
857
+ {
858
+ "from": "gpt",
859
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 1\n overall score: 1\n"
860
+ }
861
+ ]
862
+ },
863
+ {
864
+ "id": "b207095",
865
+ "images": [
866
+ "b207095_00.jpg",
867
+ "b207095_01.jpg",
868
+ "b207095_02.jpg",
869
+ "b207095_03.jpg",
870
+ "b207095_04.jpg",
871
+ "b207095_05.jpg",
872
+ "b207095_06.jpg",
873
+ "b207095_07.jpg",
874
+ "b207095_08.jpg",
875
+ "b207095_09.jpg",
876
+ "b207095_10.jpg",
877
+ "b207095_11.jpg",
878
+ "b207095_12.jpg",
879
+ "b207095_13.jpg",
880
+ "b207095_14.jpg",
881
+ "b207095_15.jpg"
882
+ ],
883
+ "conversations": [
884
+ {
885
+ "from": "human",
886
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is A person, back view, starry sky, Milky Way, desert, panorama, high definition ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
887
+ },
888
+ {
889
+ "from": "gpt",
890
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 1\n text-to-video alignment: 3\n factual consistency: 2\n overall score: 1\n"
891
+ }
892
+ ]
893
+ },
894
+ {
895
+ "id": "b703225",
896
+ "images": [
897
+ "b703225_00.jpg",
898
+ "b703225_01.jpg",
899
+ "b703225_02.jpg",
900
+ "b703225_03.jpg",
901
+ "b703225_04.jpg",
902
+ "b703225_05.jpg",
903
+ "b703225_06.jpg",
904
+ "b703225_07.jpg"
905
+ ],
906
+ "conversations": [
907
+ {
908
+ "from": "human",
909
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is a couple in a vhs store in the 1990s alternative grunge ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> \n"
910
+ },
911
+ {
912
+ "from": "gpt",
913
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 1\n"
914
+ }
915
+ ]
916
+ },
917
+ {
918
+ "id": "b206798",
919
+ "images": [
920
+ "b206798_00.jpg",
921
+ "b206798_01.jpg",
922
+ "b206798_02.jpg",
923
+ "b206798_03.jpg",
924
+ "b206798_04.jpg",
925
+ "b206798_05.jpg",
926
+ "b206798_06.jpg",
927
+ "b206798_07.jpg",
928
+ "b206798_08.jpg",
929
+ "b206798_09.jpg",
930
+ "b206798_10.jpg",
931
+ "b206798_11.jpg",
932
+ "b206798_12.jpg",
933
+ "b206798_13.jpg",
934
+ "b206798_14.jpg",
935
+ "b206798_15.jpg"
936
+ ],
937
+ "conversations": [
938
+ {
939
+ "from": "human",
940
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is long shot pushing into of lit bar in war in torn Serbia with bright yellow lights seen within and several military men milling in the background, high detail ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
941
+ },
942
+ {
943
+ "from": "gpt",
944
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 1\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 1\n"
945
+ }
946
+ ]
947
+ },
948
+ {
949
+ "id": "b203583",
950
+ "images": [
951
+ "b203583_00.jpg",
952
+ "b203583_01.jpg",
953
+ "b203583_02.jpg",
954
+ "b203583_03.jpg",
955
+ "b203583_04.jpg",
956
+ "b203583_05.jpg",
957
+ "b203583_06.jpg",
958
+ "b203583_07.jpg",
959
+ "b203583_08.jpg",
960
+ "b203583_09.jpg",
961
+ "b203583_10.jpg",
962
+ "b203583_11.jpg",
963
+ "b203583_12.jpg",
964
+ "b203583_13.jpg",
965
+ "b203583_14.jpg",
966
+ "b203583_15.jpg"
967
+ ],
968
+ "conversations": [
969
+ {
970
+ "from": "human",
971
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is red grass spreading around the normal grass turning the normal grass in red grass ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
972
+ },
973
+ {
974
+ "from": "gpt",
975
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 1\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
976
+ }
977
+ ]
978
+ },
979
+ {
980
+ "id": "b106024",
981
+ "images": [
982
+ "b106024_00.jpg",
983
+ "b106024_01.jpg",
984
+ "b106024_02.jpg",
985
+ "b106024_03.jpg",
986
+ "b106024_04.jpg",
987
+ "b106024_05.jpg",
988
+ "b106024_06.jpg",
989
+ "b106024_07.jpg",
990
+ "b106024_08.jpg",
991
+ "b106024_09.jpg",
992
+ "b106024_10.jpg",
993
+ "b106024_11.jpg",
994
+ "b106024_12.jpg",
995
+ "b106024_13.jpg",
996
+ "b106024_14.jpg",
997
+ "b106024_15.jpg"
998
+ ],
999
+ "conversations": [
1000
+ {
1001
+ "from": "human",
1002
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is a guitar and surfboard big waves ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1003
+ },
1004
+ {
1005
+ "from": "gpt",
1006
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
1007
+ }
1008
+ ]
1009
+ },
1010
+ {
1011
+ "id": "b306558",
1012
+ "images": [
1013
+ "b306558_00.jpg",
1014
+ "b306558_01.jpg",
1015
+ "b306558_02.jpg",
1016
+ "b306558_03.jpg",
1017
+ "b306558_04.jpg",
1018
+ "b306558_05.jpg",
1019
+ "b306558_06.jpg",
1020
+ "b306558_07.jpg",
1021
+ "b306558_08.jpg",
1022
+ "b306558_09.jpg",
1023
+ "b306558_10.jpg",
1024
+ "b306558_11.jpg",
1025
+ "b306558_12.jpg",
1026
+ "b306558_13.jpg",
1027
+ "b306558_14.jpg",
1028
+ "b306558_15.jpg"
1029
+ ],
1030
+ "conversations": [
1031
+ {
1032
+ "from": "human",
1033
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is stage Pyrotechnics burst at the edge of the stage while the band is playing ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1034
+ },
1035
+ {
1036
+ "from": "gpt",
1037
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 1\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
1038
+ }
1039
+ ]
1040
+ },
1041
+ {
1042
+ "id": "b803709",
1043
+ "images": [
1044
+ "b803709_00.jpg",
1045
+ "b803709_01.jpg",
1046
+ "b803709_02.jpg",
1047
+ "b803709_03.jpg",
1048
+ "b803709_04.jpg",
1049
+ "b803709_05.jpg",
1050
+ "b803709_06.jpg",
1051
+ "b803709_07.jpg",
1052
+ "b803709_08.jpg",
1053
+ "b803709_09.jpg",
1054
+ "b803709_10.jpg",
1055
+ "b803709_11.jpg",
1056
+ "b803709_12.jpg",
1057
+ "b803709_13.jpg",
1058
+ "b803709_14.jpg",
1059
+ "b803709_15.jpg",
1060
+ "b803709_16.jpg",
1061
+ "b803709_17.jpg",
1062
+ "b803709_18.jpg",
1063
+ "b803709_19.jpg",
1064
+ "b803709_20.jpg",
1065
+ "b803709_21.jpg",
1066
+ "b803709_22.jpg",
1067
+ "b803709_23.jpg"
1068
+ ],
1069
+ "conversations": [
1070
+ {
1071
+ "from": "human",
1072
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is POV from behind, person running ancient roman architecture parkour course, illustration by moebius and Ben Bauchau beautiful, complicated labrythn pool garden, ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1073
+ },
1074
+ {
1075
+ "from": "gpt",
1076
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 1\n factual consistency: 3\n overall score: 1\n"
1077
+ }
1078
+ ]
1079
+ },
1080
+ {
1081
+ "id": "b205414",
1082
+ "images": [
1083
+ "b205414_00.jpg",
1084
+ "b205414_01.jpg",
1085
+ "b205414_02.jpg",
1086
+ "b205414_03.jpg",
1087
+ "b205414_04.jpg",
1088
+ "b205414_05.jpg",
1089
+ "b205414_06.jpg",
1090
+ "b205414_07.jpg",
1091
+ "b205414_08.jpg",
1092
+ "b205414_09.jpg",
1093
+ "b205414_10.jpg",
1094
+ "b205414_11.jpg",
1095
+ "b205414_12.jpg",
1096
+ "b205414_13.jpg",
1097
+ "b205414_14.jpg",
1098
+ "b205414_15.jpg"
1099
+ ],
1100
+ "conversations": [
1101
+ {
1102
+ "from": "human",
1103
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is Use imaginative and playful animations to make it exciting for kids. ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1104
+ },
1105
+ {
1106
+ "from": "gpt",
1107
+ "value": "visual quality: 1\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 1\n"
1108
+ }
1109
+ ]
1110
+ },
1111
+ {
1112
+ "id": "b006836",
1113
+ "images": [
1114
+ "b006836_00.jpg",
1115
+ "b006836_01.jpg",
1116
+ "b006836_02.jpg",
1117
+ "b006836_03.jpg",
1118
+ "b006836_04.jpg",
1119
+ "b006836_05.jpg",
1120
+ "b006836_06.jpg",
1121
+ "b006836_07.jpg",
1122
+ "b006836_08.jpg",
1123
+ "b006836_09.jpg",
1124
+ "b006836_10.jpg",
1125
+ "b006836_11.jpg",
1126
+ "b006836_12.jpg",
1127
+ "b006836_13.jpg",
1128
+ "b006836_14.jpg",
1129
+ "b006836_15.jpg",
1130
+ "b006836_16.jpg",
1131
+ "b006836_17.jpg",
1132
+ "b006836_18.jpg",
1133
+ "b006836_19.jpg",
1134
+ "b006836_20.jpg",
1135
+ "b006836_21.jpg",
1136
+ "b006836_22.jpg",
1137
+ "b006836_23.jpg"
1138
+ ],
1139
+ "conversations": [
1140
+ {
1141
+ "from": "human",
1142
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is An otherworldly landscape unfolds before us, with towering mountains that seem to touch the sky, their peaks crowned with swirling clouds of vibrant colors, like a dream painted by Andrei Platounov ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1143
+ },
1144
+ {
1145
+ "from": "gpt",
1146
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 2\n factual consistency: 2\n overall score: 1\n"
1147
+ }
1148
+ ]
1149
+ },
1150
+ {
1151
+ "id": "b005955",
1152
+ "images": [
1153
+ "b005955_00.jpg",
1154
+ "b005955_01.jpg",
1155
+ "b005955_02.jpg",
1156
+ "b005955_03.jpg",
1157
+ "b005955_04.jpg",
1158
+ "b005955_05.jpg",
1159
+ "b005955_06.jpg",
1160
+ "b005955_07.jpg",
1161
+ "b005955_08.jpg",
1162
+ "b005955_09.jpg",
1163
+ "b005955_10.jpg",
1164
+ "b005955_11.jpg",
1165
+ "b005955_12.jpg",
1166
+ "b005955_13.jpg",
1167
+ "b005955_14.jpg",
1168
+ "b005955_15.jpg",
1169
+ "b005955_16.jpg",
1170
+ "b005955_17.jpg",
1171
+ "b005955_18.jpg",
1172
+ "b005955_19.jpg",
1173
+ "b005955_20.jpg",
1174
+ "b005955_21.jpg",
1175
+ "b005955_22.jpg",
1176
+ "b005955_23.jpg"
1177
+ ],
1178
+ "conversations": [
1179
+ {
1180
+ "from": "human",
1181
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is The king appointed a trainer for the parrot to speak. Many days went by but the parrot did not speak. Even the royal princes tried to talk with the parrot, but the parrot did not speak a word. ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1182
+ },
1183
+ {
1184
+ "from": "gpt",
1185
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
1186
+ }
1187
+ ]
1188
+ },
1189
+ {
1190
+ "id": "b803572",
1191
+ "images": [
1192
+ "b803572_00.jpg",
1193
+ "b803572_01.jpg",
1194
+ "b803572_02.jpg",
1195
+ "b803572_03.jpg",
1196
+ "b803572_04.jpg",
1197
+ "b803572_05.jpg",
1198
+ "b803572_06.jpg",
1199
+ "b803572_07.jpg",
1200
+ "b803572_08.jpg",
1201
+ "b803572_09.jpg",
1202
+ "b803572_10.jpg",
1203
+ "b803572_11.jpg",
1204
+ "b803572_12.jpg",
1205
+ "b803572_13.jpg",
1206
+ "b803572_14.jpg",
1207
+ "b803572_15.jpg",
1208
+ "b803572_16.jpg",
1209
+ "b803572_17.jpg",
1210
+ "b803572_18.jpg",
1211
+ "b803572_19.jpg",
1212
+ "b803572_20.jpg",
1213
+ "b803572_21.jpg",
1214
+ "b803572_22.jpg",
1215
+ "b803572_23.jpg"
1216
+ ],
1217
+ "conversations": [
1218
+ {
1219
+ "from": "human",
1220
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is Batman driving a hellcat,while joker is racing him with a trackhawk srt ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1221
+ },
1222
+ {
1223
+ "from": "gpt",
1224
+ "value": "visual quality: 2\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 2\n text-to-video alignment: 1\n factual consistency: 1\n overall score: 1\n"
1225
+ }
1226
+ ]
1227
+ },
1228
+ {
1229
+ "id": "b403738",
1230
+ "images": [
1231
+ "b403738_00.jpg",
1232
+ "b403738_01.jpg",
1233
+ "b403738_02.jpg",
1234
+ "b403738_03.jpg",
1235
+ "b403738_04.jpg",
1236
+ "b403738_05.jpg",
1237
+ "b403738_06.jpg",
1238
+ "b403738_07.jpg",
1239
+ "b403738_08.jpg",
1240
+ "b403738_09.jpg",
1241
+ "b403738_10.jpg",
1242
+ "b403738_11.jpg",
1243
+ "b403738_12.jpg",
1244
+ "b403738_13.jpg",
1245
+ "b403738_14.jpg",
1246
+ "b403738_15.jpg"
1247
+ ],
1248
+ "conversations": [
1249
+ {
1250
+ "from": "human",
1251
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is beautiful Sunset on a big yacht, photorealistic, Ultrarealistic, cinema, ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1252
+ },
1253
+ {
1254
+ "from": "gpt",
1255
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 1\n motion smoothness: 3\n text-to-video alignment: 2\n factual consistency: 3\n overall score: 1\n"
1256
+ }
1257
+ ]
1258
+ },
1259
+ {
1260
+ "id": "b004393",
1261
+ "images": [
1262
+ "b004393_00.jpg",
1263
+ "b004393_01.jpg",
1264
+ "b004393_02.jpg",
1265
+ "b004393_03.jpg",
1266
+ "b004393_04.jpg",
1267
+ "b004393_05.jpg",
1268
+ "b004393_06.jpg",
1269
+ "b004393_07.jpg",
1270
+ "b004393_08.jpg",
1271
+ "b004393_09.jpg",
1272
+ "b004393_10.jpg",
1273
+ "b004393_11.jpg",
1274
+ "b004393_12.jpg",
1275
+ "b004393_13.jpg",
1276
+ "b004393_14.jpg",
1277
+ "b004393_15.jpg",
1278
+ "b004393_16.jpg",
1279
+ "b004393_17.jpg",
1280
+ "b004393_18.jpg",
1281
+ "b004393_19.jpg",
1282
+ "b004393_20.jpg",
1283
+ "b004393_21.jpg",
1284
+ "b004393_22.jpg",
1285
+ "b004393_23.jpg"
1286
+ ],
1287
+ "conversations": [
1288
+ {
1289
+ "from": "human",
1290
+ "value": "\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, \nplease watch the following frames of a given video and see the text prompt for generating the video, \nthen give scores from 7 different dimensions:\n(1) visual quality, \n(2) object consistency,\n(3) dynamic degree,\n(4) motion smoothness,\n(5) text-to-video alignment,\n(6) factual consistency, \n(7) overall score\nfor each dimension, output a number from [1,2,3], in which '1' stands for 'Bad', '2' stands for 'Average', '3' stands for 'Good'.\nHere is an output example: \nvisual quality: 3\nobject consistency: 2 \ndynamic degree: 2\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this item, the text prompt is There are various small animals in the wonderland ,\nall the frames of video are as follows: \n\n<image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> <image> \n"
1291
+ },
1292
+ {
1293
+ "from": "gpt",
1294
+ "value": "visual quality: 3\n object consistency: 1\n dynamic degree: 3\n motion smoothness: 1\n text-to-video alignment: 3\n factual consistency: 2\n overall score: 1\n"
1295
+ }
1296
+ ]
1297
+ }
1298
+ ]