Hugging Face's logo

Datasets:
hexuan21
/
VideoScore_legacy

License:
Dataset card Data Studio Files
xet
Community
hexuan21 commited on
Commit
ae875e2
·
verified ·
1 Parent(s): 7055853

upload high_res_35k/test_250 's json, videos and frames

Browse files
Files changed (1) hide show
  1. data/high_res_35k/test_250/data_real_high_res.json +782 -0
data/high_res_35k/test_250/data_real_high_res.json ADDED
@@ -0,0 +1,782 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "id": "p002988",
4
+ "images": [
5
+ "p002988_00.jpg",
6
+ "p002988_01.jpg",
7
+ "p002988_02.jpg",
8
+ "p002988_03.jpg",
9
+ "p002988_04.jpg",
10
+ "p002988_05.jpg",
11
+ "p002988_06.jpg",
12
+ "p002988_07.jpg",
13
+ "p002988_08.jpg",
14
+ "p002988_09.jpg",
15
+ "p002988_10.jpg",
16
+ "p002988_11.jpg",
17
+ "p002988_12.jpg",
18
+ "p002988_13.jpg",
19
+ "p002988_14.jpg",
20
+ "p002988_15.jpg",
21
+ "p002988_16.jpg",
22
+ "p002988_17.jpg",
23
+ "p002988_18.jpg",
24
+ "p002988_19.jpg",
25
+ "p002988_20.jpg",
26
+ "p002988_21.jpg",
27
+ "p002988_22.jpg",
28
+ "p002988_23.jpg"
29
+ ],
30
+ "conversations": [
31
+ {
32
+ "from": "human",
33
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"two white planes in the grass on the side of moving plane\",\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"
34
+ },
35
+ {
36
+ "from": "gpt",
37
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
38
+ }
39
+ ]
40
+ },
41
+ {
42
+ "id": "p002932",
43
+ "images": [
44
+ "p002932_00.jpg",
45
+ "p002932_01.jpg",
46
+ "p002932_02.jpg",
47
+ "p002932_03.jpg",
48
+ "p002932_04.jpg",
49
+ "p002932_05.jpg",
50
+ "p002932_06.jpg",
51
+ "p002932_07.jpg",
52
+ "p002932_08.jpg",
53
+ "p002932_09.jpg",
54
+ "p002932_10.jpg",
55
+ "p002932_11.jpg",
56
+ "p002932_12.jpg",
57
+ "p002932_13.jpg",
58
+ "p002932_14.jpg",
59
+ "p002932_15.jpg",
60
+ "p002932_16.jpg",
61
+ "p002932_17.jpg",
62
+ "p002932_18.jpg",
63
+ "p002932_19.jpg",
64
+ "p002932_20.jpg",
65
+ "p002932_21.jpg",
66
+ "p002932_22.jpg",
67
+ "p002932_23.jpg"
68
+ ],
69
+ "conversations": [
70
+ {
71
+ "from": "human",
72
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"woman smiles lady wearing grey scarf enter frame\",\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"
73
+ },
74
+ {
75
+ "from": "gpt",
76
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
77
+ }
78
+ ]
79
+ },
80
+ {
81
+ "id": "p002661",
82
+ "images": [
83
+ "p002661_00.jpg",
84
+ "p002661_01.jpg",
85
+ "p002661_02.jpg",
86
+ "p002661_03.jpg",
87
+ "p002661_04.jpg",
88
+ "p002661_05.jpg",
89
+ "p002661_06.jpg",
90
+ "p002661_07.jpg",
91
+ "p002661_08.jpg",
92
+ "p002661_09.jpg",
93
+ "p002661_10.jpg",
94
+ "p002661_11.jpg",
95
+ "p002661_12.jpg",
96
+ "p002661_13.jpg",
97
+ "p002661_14.jpg",
98
+ "p002661_15.jpg",
99
+ "p002661_16.jpg",
100
+ "p002661_17.jpg",
101
+ "p002661_18.jpg",
102
+ "p002661_19.jpg",
103
+ "p002661_20.jpg",
104
+ "p002661_21.jpg",
105
+ "p002661_22.jpg",
106
+ "p002661_23.jpg"
107
+ ],
108
+ "conversations": [
109
+ {
110
+ "from": "human",
111
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"person sitting in background with legs crossed\",\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"
112
+ },
113
+ {
114
+ "from": "gpt",
115
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
116
+ }
117
+ ]
118
+ },
119
+ {
120
+ "id": "p000814",
121
+ "images": [
122
+ "p000814_00.jpg",
123
+ "p000814_01.jpg",
124
+ "p000814_02.jpg",
125
+ "p000814_03.jpg",
126
+ "p000814_04.jpg",
127
+ "p000814_05.jpg",
128
+ "p000814_06.jpg",
129
+ "p000814_07.jpg",
130
+ "p000814_08.jpg",
131
+ "p000814_09.jpg",
132
+ "p000814_10.jpg",
133
+ "p000814_11.jpg",
134
+ "p000814_12.jpg",
135
+ "p000814_13.jpg",
136
+ "p000814_14.jpg",
137
+ "p000814_15.jpg",
138
+ "p000814_16.jpg",
139
+ "p000814_17.jpg",
140
+ "p000814_18.jpg",
141
+ "p000814_19.jpg",
142
+ "p000814_20.jpg",
143
+ "p000814_21.jpg",
144
+ "p000814_22.jpg",
145
+ "p000814_23.jpg"
146
+ ],
147
+ "conversations": [
148
+ {
149
+ "from": "human",
150
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"Five girls enter the stage and start walking\",\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"
151
+ },
152
+ {
153
+ "from": "gpt",
154
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
155
+ }
156
+ ]
157
+ },
158
+ {
159
+ "id": "p001728",
160
+ "images": [
161
+ "p001728_00.jpg",
162
+ "p001728_01.jpg",
163
+ "p001728_02.jpg",
164
+ "p001728_03.jpg",
165
+ "p001728_04.jpg",
166
+ "p001728_05.jpg",
167
+ "p001728_06.jpg",
168
+ "p001728_07.jpg",
169
+ "p001728_08.jpg",
170
+ "p001728_09.jpg",
171
+ "p001728_10.jpg",
172
+ "p001728_11.jpg",
173
+ "p001728_12.jpg",
174
+ "p001728_13.jpg",
175
+ "p001728_14.jpg",
176
+ "p001728_15.jpg",
177
+ "p001728_16.jpg",
178
+ "p001728_17.jpg",
179
+ "p001728_18.jpg",
180
+ "p001728_19.jpg",
181
+ "p001728_20.jpg",
182
+ "p001728_21.jpg",
183
+ "p001728_22.jpg",
184
+ "p001728_23.jpg"
185
+ ],
186
+ "conversations": [
187
+ {
188
+ "from": "human",
189
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"cat rubs up against porch\",\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"
190
+ },
191
+ {
192
+ "from": "gpt",
193
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
194
+ }
195
+ ]
196
+ },
197
+ {
198
+ "id": "p001470",
199
+ "images": [
200
+ "p001470_00.jpg",
201
+ "p001470_01.jpg",
202
+ "p001470_02.jpg",
203
+ "p001470_03.jpg",
204
+ "p001470_04.jpg",
205
+ "p001470_05.jpg",
206
+ "p001470_06.jpg",
207
+ "p001470_07.jpg",
208
+ "p001470_08.jpg",
209
+ "p001470_09.jpg",
210
+ "p001470_10.jpg",
211
+ "p001470_11.jpg",
212
+ "p001470_12.jpg",
213
+ "p001470_13.jpg",
214
+ "p001470_14.jpg",
215
+ "p001470_15.jpg",
216
+ "p001470_16.jpg",
217
+ "p001470_17.jpg",
218
+ "p001470_18.jpg",
219
+ "p001470_19.jpg",
220
+ "p001470_20.jpg",
221
+ "p001470_21.jpg",
222
+ "p001470_22.jpg",
223
+ "p001470_23.jpg"
224
+ ],
225
+ "conversations": [
226
+ {
227
+ "from": "human",
228
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"a child swings a bat at ball.\",\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"
229
+ },
230
+ {
231
+ "from": "gpt",
232
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
233
+ }
234
+ ]
235
+ },
236
+ {
237
+ "id": "p002413",
238
+ "images": [
239
+ "p002413_00.jpg",
240
+ "p002413_01.jpg",
241
+ "p002413_02.jpg",
242
+ "p002413_03.jpg",
243
+ "p002413_04.jpg",
244
+ "p002413_05.jpg",
245
+ "p002413_06.jpg",
246
+ "p002413_07.jpg",
247
+ "p002413_08.jpg",
248
+ "p002413_09.jpg",
249
+ "p002413_10.jpg",
250
+ "p002413_11.jpg",
251
+ "p002413_12.jpg",
252
+ "p002413_13.jpg",
253
+ "p002413_14.jpg",
254
+ "p002413_15.jpg",
255
+ "p002413_16.jpg",
256
+ "p002413_17.jpg",
257
+ "p002413_18.jpg",
258
+ "p002413_19.jpg",
259
+ "p002413_20.jpg",
260
+ "p002413_21.jpg",
261
+ "p002413_22.jpg",
262
+ "p002413_23.jpg"
263
+ ],
264
+ "conversations": [
265
+ {
266
+ "from": "human",
267
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"a black car goesthrough the traffic and a red bus goes by\",\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"
268
+ },
269
+ {
270
+ "from": "gpt",
271
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
272
+ }
273
+ ]
274
+ },
275
+ {
276
+ "id": "p000383",
277
+ "images": [
278
+ "p000383_00.jpg",
279
+ "p000383_01.jpg",
280
+ "p000383_02.jpg",
281
+ "p000383_03.jpg",
282
+ "p000383_04.jpg",
283
+ "p000383_05.jpg",
284
+ "p000383_06.jpg",
285
+ "p000383_07.jpg",
286
+ "p000383_08.jpg",
287
+ "p000383_09.jpg",
288
+ "p000383_10.jpg",
289
+ "p000383_11.jpg",
290
+ "p000383_12.jpg",
291
+ "p000383_13.jpg",
292
+ "p000383_14.jpg",
293
+ "p000383_15.jpg",
294
+ "p000383_16.jpg",
295
+ "p000383_17.jpg",
296
+ "p000383_18.jpg",
297
+ "p000383_19.jpg",
298
+ "p000383_20.jpg",
299
+ "p000383_21.jpg",
300
+ "p000383_22.jpg",
301
+ "p000383_23.jpg"
302
+ ],
303
+ "conversations": [
304
+ {
305
+ "from": "human",
306
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"a car does a turn and manages to drive away.\",\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"
307
+ },
308
+ {
309
+ "from": "gpt",
310
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
311
+ }
312
+ ]
313
+ },
314
+ {
315
+ "id": "p000456",
316
+ "images": [
317
+ "p000456_00.jpg",
318
+ "p000456_01.jpg",
319
+ "p000456_02.jpg",
320
+ "p000456_03.jpg",
321
+ "p000456_04.jpg",
322
+ "p000456_05.jpg",
323
+ "p000456_06.jpg",
324
+ "p000456_07.jpg",
325
+ "p000456_08.jpg",
326
+ "p000456_09.jpg",
327
+ "p000456_10.jpg",
328
+ "p000456_11.jpg",
329
+ "p000456_12.jpg",
330
+ "p000456_13.jpg",
331
+ "p000456_14.jpg",
332
+ "p000456_15.jpg",
333
+ "p000456_16.jpg",
334
+ "p000456_17.jpg",
335
+ "p000456_18.jpg",
336
+ "p000456_19.jpg",
337
+ "p000456_20.jpg",
338
+ "p000456_21.jpg",
339
+ "p000456_22.jpg",
340
+ "p000456_23.jpg"
341
+ ],
342
+ "conversations": [
343
+ {
344
+ "from": "human",
345
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"man in the hat turns camera towards himself.\",\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"
346
+ },
347
+ {
348
+ "from": "gpt",
349
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
350
+ }
351
+ ]
352
+ },
353
+ {
354
+ "id": "p000653",
355
+ "images": [
356
+ "p000653_00.jpg",
357
+ "p000653_01.jpg",
358
+ "p000653_02.jpg",
359
+ "p000653_03.jpg",
360
+ "p000653_04.jpg",
361
+ "p000653_05.jpg",
362
+ "p000653_06.jpg",
363
+ "p000653_07.jpg",
364
+ "p000653_08.jpg",
365
+ "p000653_09.jpg",
366
+ "p000653_10.jpg",
367
+ "p000653_11.jpg",
368
+ "p000653_12.jpg",
369
+ "p000653_13.jpg",
370
+ "p000653_14.jpg",
371
+ "p000653_15.jpg",
372
+ "p000653_16.jpg",
373
+ "p000653_17.jpg",
374
+ "p000653_18.jpg",
375
+ "p000653_19.jpg",
376
+ "p000653_20.jpg",
377
+ "p000653_21.jpg",
378
+ "p000653_22.jpg",
379
+ "p000653_23.jpg"
380
+ ],
381
+ "conversations": [
382
+ {
383
+ "from": "human",
384
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"a man in a green shirt and a man in a blue cap behind a microphone and a laptop.\",\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"
385
+ },
386
+ {
387
+ "from": "gpt",
388
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
389
+ }
390
+ ]
391
+ },
392
+ {
393
+ "id": "p002859",
394
+ "images": [
395
+ "p002859_00.jpg",
396
+ "p002859_01.jpg",
397
+ "p002859_02.jpg",
398
+ "p002859_03.jpg",
399
+ "p002859_04.jpg",
400
+ "p002859_05.jpg",
401
+ "p002859_06.jpg",
402
+ "p002859_07.jpg",
403
+ "p002859_08.jpg",
404
+ "p002859_09.jpg",
405
+ "p002859_10.jpg",
406
+ "p002859_11.jpg",
407
+ "p002859_12.jpg",
408
+ "p002859_13.jpg",
409
+ "p002859_14.jpg",
410
+ "p002859_15.jpg",
411
+ "p002859_16.jpg",
412
+ "p002859_17.jpg",
413
+ "p002859_18.jpg",
414
+ "p002859_19.jpg",
415
+ "p002859_20.jpg",
416
+ "p002859_21.jpg",
417
+ "p002859_22.jpg",
418
+ "p002859_23.jpg"
419
+ ],
420
+ "conversations": [
421
+ {
422
+ "from": "human",
423
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"you can see fins poking up out of the water\",\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"
424
+ },
425
+ {
426
+ "from": "gpt",
427
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
428
+ }
429
+ ]
430
+ },
431
+ {
432
+ "id": "p001408",
433
+ "images": [
434
+ "p001408_00.jpg",
435
+ "p001408_01.jpg",
436
+ "p001408_02.jpg",
437
+ "p001408_03.jpg",
438
+ "p001408_04.jpg",
439
+ "p001408_05.jpg",
440
+ "p001408_06.jpg",
441
+ "p001408_07.jpg",
442
+ "p001408_08.jpg",
443
+ "p001408_09.jpg",
444
+ "p001408_10.jpg",
445
+ "p001408_11.jpg",
446
+ "p001408_12.jpg",
447
+ "p001408_13.jpg",
448
+ "p001408_14.jpg",
449
+ "p001408_15.jpg",
450
+ "p001408_16.jpg",
451
+ "p001408_17.jpg",
452
+ "p001408_18.jpg",
453
+ "p001408_19.jpg",
454
+ "p001408_20.jpg",
455
+ "p001408_21.jpg",
456
+ "p001408_22.jpg",
457
+ "p001408_23.jpg"
458
+ ],
459
+ "conversations": [
460
+ {
461
+ "from": "human",
462
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"girl excitedly talking and waving\",\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"
463
+ },
464
+ {
465
+ "from": "gpt",
466
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
467
+ }
468
+ ]
469
+ },
470
+ {
471
+ "id": "p001731",
472
+ "images": [
473
+ "p001731_00.jpg",
474
+ "p001731_01.jpg",
475
+ "p001731_02.jpg",
476
+ "p001731_03.jpg",
477
+ "p001731_04.jpg",
478
+ "p001731_05.jpg",
479
+ "p001731_06.jpg",
480
+ "p001731_07.jpg",
481
+ "p001731_08.jpg",
482
+ "p001731_09.jpg",
483
+ "p001731_10.jpg",
484
+ "p001731_11.jpg",
485
+ "p001731_12.jpg",
486
+ "p001731_13.jpg",
487
+ "p001731_14.jpg",
488
+ "p001731_15.jpg",
489
+ "p001731_16.jpg",
490
+ "p001731_17.jpg",
491
+ "p001731_18.jpg",
492
+ "p001731_19.jpg",
493
+ "p001731_20.jpg",
494
+ "p001731_21.jpg",
495
+ "p001731_22.jpg",
496
+ "p001731_23.jpg"
497
+ ],
498
+ "conversations": [
499
+ {
500
+ "from": "human",
501
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"the women stares at her food while chewing, about to pick up another bite.\",\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"
502
+ },
503
+ {
504
+ "from": "gpt",
505
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
506
+ }
507
+ ]
508
+ },
509
+ {
510
+ "id": "p001550",
511
+ "images": [
512
+ "p001550_00.jpg",
513
+ "p001550_01.jpg",
514
+ "p001550_02.jpg",
515
+ "p001550_03.jpg",
516
+ "p001550_04.jpg",
517
+ "p001550_05.jpg",
518
+ "p001550_06.jpg",
519
+ "p001550_07.jpg",
520
+ "p001550_08.jpg",
521
+ "p001550_09.jpg",
522
+ "p001550_10.jpg",
523
+ "p001550_11.jpg",
524
+ "p001550_12.jpg",
525
+ "p001550_13.jpg",
526
+ "p001550_14.jpg",
527
+ "p001550_15.jpg",
528
+ "p001550_16.jpg",
529
+ "p001550_17.jpg",
530
+ "p001550_18.jpg",
531
+ "p001550_19.jpg",
532
+ "p001550_20.jpg",
533
+ "p001550_21.jpg",
534
+ "p001550_22.jpg",
535
+ "p001550_23.jpg"
536
+ ],
537
+ "conversations": [
538
+ {
539
+ "from": "human",
540
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"the kid in blue dribbles the ball.\",\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"
541
+ },
542
+ {
543
+ "from": "gpt",
544
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
545
+ }
546
+ ]
547
+ },
548
+ {
549
+ "id": "p002298",
550
+ "images": [
551
+ "p002298_00.jpg",
552
+ "p002298_01.jpg",
553
+ "p002298_02.jpg",
554
+ "p002298_03.jpg",
555
+ "p002298_04.jpg",
556
+ "p002298_05.jpg",
557
+ "p002298_06.jpg",
558
+ "p002298_07.jpg",
559
+ "p002298_08.jpg",
560
+ "p002298_09.jpg",
561
+ "p002298_10.jpg",
562
+ "p002298_11.jpg",
563
+ "p002298_12.jpg",
564
+ "p002298_13.jpg",
565
+ "p002298_14.jpg",
566
+ "p002298_15.jpg",
567
+ "p002298_16.jpg",
568
+ "p002298_17.jpg",
569
+ "p002298_18.jpg",
570
+ "p002298_19.jpg",
571
+ "p002298_20.jpg",
572
+ "p002298_21.jpg",
573
+ "p002298_22.jpg",
574
+ "p002298_23.jpg"
575
+ ],
576
+ "conversations": [
577
+ {
578
+ "from": "human",
579
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"a woman wearing a red shirt appears to get on the escalator\",\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"
580
+ },
581
+ {
582
+ "from": "gpt",
583
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
584
+ }
585
+ ]
586
+ },
587
+ {
588
+ "id": "p000177",
589
+ "images": [
590
+ "p000177_00.jpg",
591
+ "p000177_01.jpg",
592
+ "p000177_02.jpg",
593
+ "p000177_03.jpg",
594
+ "p000177_04.jpg",
595
+ "p000177_05.jpg",
596
+ "p000177_06.jpg",
597
+ "p000177_07.jpg",
598
+ "p000177_08.jpg",
599
+ "p000177_09.jpg",
600
+ "p000177_10.jpg",
601
+ "p000177_11.jpg",
602
+ "p000177_12.jpg",
603
+ "p000177_13.jpg",
604
+ "p000177_14.jpg",
605
+ "p000177_15.jpg",
606
+ "p000177_16.jpg",
607
+ "p000177_17.jpg",
608
+ "p000177_18.jpg",
609
+ "p000177_19.jpg",
610
+ "p000177_20.jpg",
611
+ "p000177_21.jpg",
612
+ "p000177_22.jpg",
613
+ "p000177_23.jpg"
614
+ ],
615
+ "conversations": [
616
+ {
617
+ "from": "human",
618
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"models approach end of catwalk, pause and turn around\",\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"
619
+ },
620
+ {
621
+ "from": "gpt",
622
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
623
+ }
624
+ ]
625
+ },
626
+ {
627
+ "id": "p001556",
628
+ "images": [
629
+ "p001556_00.jpg",
630
+ "p001556_01.jpg",
631
+ "p001556_02.jpg",
632
+ "p001556_03.jpg",
633
+ "p001556_04.jpg",
634
+ "p001556_05.jpg",
635
+ "p001556_06.jpg",
636
+ "p001556_07.jpg",
637
+ "p001556_08.jpg",
638
+ "p001556_09.jpg",
639
+ "p001556_10.jpg",
640
+ "p001556_11.jpg",
641
+ "p001556_12.jpg",
642
+ "p001556_13.jpg",
643
+ "p001556_14.jpg",
644
+ "p001556_15.jpg",
645
+ "p001556_16.jpg",
646
+ "p001556_17.jpg",
647
+ "p001556_18.jpg",
648
+ "p001556_19.jpg",
649
+ "p001556_20.jpg",
650
+ "p001556_21.jpg",
651
+ "p001556_22.jpg",
652
+ "p001556_23.jpg"
653
+ ],
654
+ "conversations": [
655
+ {
656
+ "from": "human",
657
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"the man eats the corn.\",\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"
658
+ },
659
+ {
660
+ "from": "gpt",
661
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
662
+ }
663
+ ]
664
+ },
665
+ {
666
+ "id": "p001083",
667
+ "images": [
668
+ "p001083_00.jpg",
669
+ "p001083_01.jpg",
670
+ "p001083_02.jpg",
671
+ "p001083_03.jpg",
672
+ "p001083_04.jpg",
673
+ "p001083_05.jpg",
674
+ "p001083_06.jpg",
675
+ "p001083_07.jpg",
676
+ "p001083_08.jpg",
677
+ "p001083_09.jpg",
678
+ "p001083_10.jpg",
679
+ "p001083_11.jpg",
680
+ "p001083_12.jpg",
681
+ "p001083_13.jpg",
682
+ "p001083_14.jpg",
683
+ "p001083_15.jpg",
684
+ "p001083_16.jpg",
685
+ "p001083_17.jpg",
686
+ "p001083_18.jpg",
687
+ "p001083_19.jpg",
688
+ "p001083_20.jpg",
689
+ "p001083_21.jpg",
690
+ "p001083_22.jpg",
691
+ "p001083_23.jpg"
692
+ ],
693
+ "conversations": [
694
+ {
695
+ "from": "human",
696
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"A man covers his head with something white.\",\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"
697
+ },
698
+ {
699
+ "from": "gpt",
700
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
701
+ }
702
+ ]
703
+ },
704
+ {
705
+ "id": "p001839",
706
+ "images": [
707
+ "p001839_00.jpg",
708
+ "p001839_01.jpg",
709
+ "p001839_02.jpg",
710
+ "p001839_03.jpg",
711
+ "p001839_04.jpg",
712
+ "p001839_05.jpg",
713
+ "p001839_06.jpg",
714
+ "p001839_07.jpg",
715
+ "p001839_08.jpg",
716
+ "p001839_09.jpg",
717
+ "p001839_10.jpg",
718
+ "p001839_11.jpg",
719
+ "p001839_12.jpg",
720
+ "p001839_13.jpg",
721
+ "p001839_14.jpg",
722
+ "p001839_15.jpg",
723
+ "p001839_16.jpg",
724
+ "p001839_17.jpg",
725
+ "p001839_18.jpg",
726
+ "p001839_19.jpg",
727
+ "p001839_20.jpg",
728
+ "p001839_21.jpg",
729
+ "p001839_22.jpg",
730
+ "p001839_23.jpg"
731
+ ],
732
+ "conversations": [
733
+ {
734
+ "from": "human",
735
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"A woman looks out the window\",\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"
736
+ },
737
+ {
738
+ "from": "gpt",
739
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
740
+ }
741
+ ]
742
+ },
743
+ {
744
+ "id": "p000881",
745
+ "images": [
746
+ "p000881_00.jpg",
747
+ "p000881_01.jpg",
748
+ "p000881_02.jpg",
749
+ "p000881_03.jpg",
750
+ "p000881_04.jpg",
751
+ "p000881_05.jpg",
752
+ "p000881_06.jpg",
753
+ "p000881_07.jpg",
754
+ "p000881_08.jpg",
755
+ "p000881_09.jpg",
756
+ "p000881_10.jpg",
757
+ "p000881_11.jpg",
758
+ "p000881_12.jpg",
759
+ "p000881_13.jpg",
760
+ "p000881_14.jpg",
761
+ "p000881_15.jpg",
762
+ "p000881_16.jpg",
763
+ "p000881_17.jpg",
764
+ "p000881_18.jpg",
765
+ "p000881_19.jpg",
766
+ "p000881_20.jpg",
767
+ "p000881_21.jpg",
768
+ "p000881_22.jpg",
769
+ "p000881_23.jpg"
770
+ ],
771
+ "conversations": [
772
+ {
773
+ "from": "human",
774
+ "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: the quality of the video in terms of clearness, resolution, brightness, and color\n(2) object consistency, the consistency of objects or humans in video\n(3) dynamic degree, the degree of dynamic changes\n(4) motion smoothness, the smoothness of motion or movements\n(5) text-to-video alignment, the alignment between the text prompt and the video content\n(6) factual consistency, the consistency of the video content with the common-sense and factual knowledge\n(7) overall score, the overall quality of the video\nfor each dimension, output a number from [1,2,3,4], \nin which '1' is 'Bad', '2' is 'Average', '3' is 'Good', '4' is 'Perfect'\nHere is an output example: \nvisual quality: 3\nobject consistency: 4\ndynamic degree: 4\nmotion smoothness: 1\ntext-to-video alignment: 1\nfactual consistency: 2\noverall score: 1\n\nFor this video, the text prompt is \"Man says words through car window while he drives by.\",\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"
775
+ },
776
+ {
777
+ "from": "gpt",
778
+ "value": "visual quality: 4\n object consistency: 4\n dynamic degree: 4\n motion smoothness: 4\n text-to-video alignment: 4\n factual consistency: 4\n overall score: 4\n"
779
+ }
780
+ ]
781
+ }
782
+ ]