Hugging Face's logo

Datasets:
hexuan21
/
VideoScore_legacy

License:
Dataset card Data Studio Files
xet
Community
hexuan21 commited on
Commit
af1cec8
·
verified ·
1 Parent(s): 713eec7

upload subset_20k/test_100 's json, videos and frames

Browse files
Files changed (1) hide show
  1. data/subset_20k/test_100/data_real.json +742 -0
data/subset_20k/test_100/data_real.json ADDED
@@ -0,0 +1,742 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "id": "r000619",
4
+ "images": [
5
+ "r000619_00.jpg",
6
+ "r000619_01.jpg",
7
+ "r000619_02.jpg",
8
+ "r000619_03.jpg",
9
+ "r000619_04.jpg",
10
+ "r000619_05.jpg",
11
+ "r000619_06.jpg",
12
+ "r000619_07.jpg",
13
+ "r000619_08.jpg",
14
+ "r000619_09.jpg",
15
+ "r000619_10.jpg",
16
+ "r000619_11.jpg",
17
+ "r000619_12.jpg",
18
+ "r000619_13.jpg",
19
+ "r000619_14.jpg",
20
+ "r000619_15.jpg",
21
+ "r000619_16.jpg",
22
+ "r000619_17.jpg",
23
+ "r000619_18.jpg",
24
+ "r000619_19.jpg",
25
+ "r000619_20.jpg",
26
+ "r000619_21.jpg",
27
+ "r000619_22.jpg",
28
+ "r000619_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, \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 elephant uses its trunk to hand something to a man.,\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: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
38
+ }
39
+ ]
40
+ },
41
+ {
42
+ "id": "r002348",
43
+ "images": [
44
+ "r002348_00.jpg",
45
+ "r002348_01.jpg",
46
+ "r002348_02.jpg",
47
+ "r002348_03.jpg",
48
+ "r002348_04.jpg",
49
+ "r002348_05.jpg",
50
+ "r002348_06.jpg",
51
+ "r002348_07.jpg",
52
+ "r002348_08.jpg",
53
+ "r002348_09.jpg",
54
+ "r002348_10.jpg",
55
+ "r002348_11.jpg",
56
+ "r002348_12.jpg",
57
+ "r002348_13.jpg",
58
+ "r002348_14.jpg",
59
+ "r002348_15.jpg",
60
+ "r002348_16.jpg",
61
+ "r002348_17.jpg",
62
+ "r002348_18.jpg",
63
+ "r002348_19.jpg",
64
+ "r002348_20.jpg",
65
+ "r002348_21.jpg",
66
+ "r002348_22.jpg",
67
+ "r002348_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, \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 boy climbs the step.,\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: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
77
+ }
78
+ ]
79
+ },
80
+ {
81
+ "id": "r102970",
82
+ "images": [
83
+ "r102970_00.jpg",
84
+ "r102970_01.jpg",
85
+ "r102970_02.jpg",
86
+ "r102970_03.jpg",
87
+ "r102970_04.jpg",
88
+ "r102970_05.jpg",
89
+ "r102970_06.jpg",
90
+ "r102970_07.jpg",
91
+ "r102970_08.jpg",
92
+ "r102970_09.jpg",
93
+ "r102970_10.jpg",
94
+ "r102970_11.jpg",
95
+ "r102970_12.jpg",
96
+ "r102970_13.jpg",
97
+ "r102970_14.jpg",
98
+ "r102970_15.jpg"
99
+ ],
100
+ "conversations": [
101
+ {
102
+ "from": "human",
103
+ "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 lego character in a video game.,\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"
104
+ },
105
+ {
106
+ "from": "gpt",
107
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
108
+ }
109
+ ]
110
+ },
111
+ {
112
+ "id": "r003062",
113
+ "images": [
114
+ "r003062_00.jpg",
115
+ "r003062_01.jpg",
116
+ "r003062_02.jpg",
117
+ "r003062_03.jpg",
118
+ "r003062_04.jpg",
119
+ "r003062_05.jpg",
120
+ "r003062_06.jpg",
121
+ "r003062_07.jpg",
122
+ "r003062_08.jpg",
123
+ "r003062_09.jpg",
124
+ "r003062_10.jpg",
125
+ "r003062_11.jpg",
126
+ "r003062_12.jpg",
127
+ "r003062_13.jpg",
128
+ "r003062_14.jpg",
129
+ "r003062_15.jpg",
130
+ "r003062_16.jpg",
131
+ "r003062_17.jpg",
132
+ "r003062_18.jpg",
133
+ "r003062_19.jpg",
134
+ "r003062_20.jpg",
135
+ "r003062_21.jpg",
136
+ "r003062_22.jpg",
137
+ "r003062_23.jpg"
138
+ ],
139
+ "conversations": [
140
+ {
141
+ "from": "human",
142
+ "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 three front performers get up and clap there hands while kicking their legs in the air.,\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"
143
+ },
144
+ {
145
+ "from": "gpt",
146
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
147
+ }
148
+ ]
149
+ },
150
+ {
151
+ "id": "r101731",
152
+ "images": [
153
+ "r101731_00.jpg",
154
+ "r101731_01.jpg",
155
+ "r101731_02.jpg",
156
+ "r101731_03.jpg",
157
+ "r101731_04.jpg",
158
+ "r101731_05.jpg",
159
+ "r101731_06.jpg",
160
+ "r101731_07.jpg",
161
+ "r101731_08.jpg",
162
+ "r101731_09.jpg",
163
+ "r101731_10.jpg",
164
+ "r101731_11.jpg",
165
+ "r101731_12.jpg",
166
+ "r101731_13.jpg",
167
+ "r101731_14.jpg",
168
+ "r101731_15.jpg",
169
+ "r101731_16.jpg",
170
+ "r101731_17.jpg",
171
+ "r101731_18.jpg",
172
+ "r101731_19.jpg",
173
+ "r101731_20.jpg",
174
+ "r101731_21.jpg",
175
+ "r101731_22.jpg",
176
+ "r101731_23.jpg"
177
+ ],
178
+ "conversations": [
179
+ {
180
+ "from": "human",
181
+ "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 woman walking in a kitchen holding a paper.,\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"
182
+ },
183
+ {
184
+ "from": "gpt",
185
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
186
+ }
187
+ ]
188
+ },
189
+ {
190
+ "id": "r004106",
191
+ "images": [
192
+ "r004106_00.jpg",
193
+ "r004106_01.jpg",
194
+ "r004106_02.jpg",
195
+ "r004106_03.jpg",
196
+ "r004106_04.jpg",
197
+ "r004106_05.jpg",
198
+ "r004106_06.jpg",
199
+ "r004106_07.jpg",
200
+ "r004106_08.jpg",
201
+ "r004106_09.jpg",
202
+ "r004106_10.jpg",
203
+ "r004106_11.jpg",
204
+ "r004106_12.jpg",
205
+ "r004106_13.jpg",
206
+ "r004106_14.jpg",
207
+ "r004106_15.jpg",
208
+ "r004106_16.jpg",
209
+ "r004106_17.jpg",
210
+ "r004106_18.jpg",
211
+ "r004106_19.jpg",
212
+ "r004106_20.jpg",
213
+ "r004106_21.jpg",
214
+ "r004106_22.jpg",
215
+ "r004106_23.jpg"
216
+ ],
217
+ "conversations": [
218
+ {
219
+ "from": "human",
220
+ "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 zooms in on the japanese temple,\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"
221
+ },
222
+ {
223
+ "from": "gpt",
224
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
225
+ }
226
+ ]
227
+ },
228
+ {
229
+ "id": "r002360",
230
+ "images": [
231
+ "r002360_00.jpg",
232
+ "r002360_01.jpg",
233
+ "r002360_02.jpg",
234
+ "r002360_03.jpg",
235
+ "r002360_04.jpg",
236
+ "r002360_05.jpg",
237
+ "r002360_06.jpg",
238
+ "r002360_07.jpg",
239
+ "r002360_08.jpg",
240
+ "r002360_09.jpg",
241
+ "r002360_10.jpg",
242
+ "r002360_11.jpg",
243
+ "r002360_12.jpg",
244
+ "r002360_13.jpg",
245
+ "r002360_14.jpg",
246
+ "r002360_15.jpg",
247
+ "r002360_16.jpg",
248
+ "r002360_17.jpg",
249
+ "r002360_18.jpg",
250
+ "r002360_19.jpg",
251
+ "r002360_20.jpg",
252
+ "r002360_21.jpg",
253
+ "r002360_22.jpg",
254
+ "r002360_23.jpg"
255
+ ],
256
+ "conversations": [
257
+ {
258
+ "from": "human",
259
+ "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 women open their fans,\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"
260
+ },
261
+ {
262
+ "from": "gpt",
263
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
264
+ }
265
+ ]
266
+ },
267
+ {
268
+ "id": "r103737",
269
+ "images": [
270
+ "r103737_00.jpg",
271
+ "r103737_01.jpg",
272
+ "r103737_02.jpg",
273
+ "r103737_03.jpg",
274
+ "r103737_04.jpg",
275
+ "r103737_05.jpg",
276
+ "r103737_06.jpg",
277
+ "r103737_07.jpg",
278
+ "r103737_08.jpg",
279
+ "r103737_09.jpg",
280
+ "r103737_10.jpg",
281
+ "r103737_11.jpg",
282
+ "r103737_12.jpg",
283
+ "r103737_13.jpg",
284
+ "r103737_14.jpg",
285
+ "r103737_15.jpg"
286
+ ],
287
+ "conversations": [
288
+ {
289
+ "from": "human",
290
+ "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 close up of a woman's hands holding a piece of paper.,\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"
291
+ },
292
+ {
293
+ "from": "gpt",
294
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
295
+ }
296
+ ]
297
+ },
298
+ {
299
+ "id": "r003408",
300
+ "images": [
301
+ "r003408_00.jpg",
302
+ "r003408_01.jpg",
303
+ "r003408_02.jpg",
304
+ "r003408_03.jpg",
305
+ "r003408_04.jpg",
306
+ "r003408_05.jpg",
307
+ "r003408_06.jpg",
308
+ "r003408_07.jpg",
309
+ "r003408_08.jpg",
310
+ "r003408_09.jpg",
311
+ "r003408_10.jpg",
312
+ "r003408_11.jpg",
313
+ "r003408_12.jpg",
314
+ "r003408_13.jpg",
315
+ "r003408_14.jpg",
316
+ "r003408_15.jpg",
317
+ "r003408_16.jpg",
318
+ "r003408_17.jpg",
319
+ "r003408_18.jpg",
320
+ "r003408_19.jpg",
321
+ "r003408_20.jpg",
322
+ "r003408_21.jpg",
323
+ "r003408_22.jpg",
324
+ "r003408_23.jpg"
325
+ ],
326
+ "conversations": [
327
+ {
328
+ "from": "human",
329
+ "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 moves down under a rail or pole.,\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"
330
+ },
331
+ {
332
+ "from": "gpt",
333
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
334
+ }
335
+ ]
336
+ },
337
+ {
338
+ "id": "r100826",
339
+ "images": [
340
+ "r100826_00.jpg",
341
+ "r100826_01.jpg",
342
+ "r100826_02.jpg",
343
+ "r100826_03.jpg",
344
+ "r100826_04.jpg",
345
+ "r100826_05.jpg",
346
+ "r100826_06.jpg",
347
+ "r100826_07.jpg",
348
+ "r100826_08.jpg",
349
+ "r100826_09.jpg",
350
+ "r100826_10.jpg",
351
+ "r100826_11.jpg",
352
+ "r100826_12.jpg",
353
+ "r100826_13.jpg",
354
+ "r100826_14.jpg",
355
+ "r100826_15.jpg"
356
+ ],
357
+ "conversations": [
358
+ {
359
+ "from": "human",
360
+ "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 bunch of tools on a wooden table.,\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"
361
+ },
362
+ {
363
+ "from": "gpt",
364
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
365
+ }
366
+ ]
367
+ },
368
+ {
369
+ "id": "r104399",
370
+ "images": [
371
+ "r104399_00.jpg",
372
+ "r104399_01.jpg",
373
+ "r104399_02.jpg",
374
+ "r104399_03.jpg",
375
+ "r104399_04.jpg",
376
+ "r104399_05.jpg",
377
+ "r104399_06.jpg",
378
+ "r104399_07.jpg",
379
+ "r104399_08.jpg",
380
+ "r104399_09.jpg",
381
+ "r104399_10.jpg",
382
+ "r104399_11.jpg",
383
+ "r104399_12.jpg",
384
+ "r104399_13.jpg",
385
+ "r104399_14.jpg",
386
+ "r104399_15.jpg",
387
+ "r104399_16.jpg",
388
+ "r104399_17.jpg",
389
+ "r104399_18.jpg",
390
+ "r104399_19.jpg",
391
+ "r104399_20.jpg",
392
+ "r104399_21.jpg",
393
+ "r104399_22.jpg",
394
+ "r104399_23.jpg"
395
+ ],
396
+ "conversations": [
397
+ {
398
+ "from": "human",
399
+ "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 swimming in the dark 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"
400
+ },
401
+ {
402
+ "from": "gpt",
403
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
404
+ }
405
+ ]
406
+ },
407
+ {
408
+ "id": "r003185",
409
+ "images": [
410
+ "r003185_00.jpg",
411
+ "r003185_01.jpg",
412
+ "r003185_02.jpg",
413
+ "r003185_03.jpg",
414
+ "r003185_04.jpg",
415
+ "r003185_05.jpg",
416
+ "r003185_06.jpg",
417
+ "r003185_07.jpg",
418
+ "r003185_08.jpg",
419
+ "r003185_09.jpg",
420
+ "r003185_10.jpg",
421
+ "r003185_11.jpg",
422
+ "r003185_12.jpg",
423
+ "r003185_13.jpg",
424
+ "r003185_14.jpg",
425
+ "r003185_15.jpg",
426
+ "r003185_16.jpg",
427
+ "r003185_17.jpg",
428
+ "r003185_18.jpg",
429
+ "r003185_19.jpg",
430
+ "r003185_20.jpg",
431
+ "r003185_21.jpg",
432
+ "r003185_22.jpg",
433
+ "r003185_23.jpg"
434
+ ],
435
+ "conversations": [
436
+ {
437
+ "from": "human",
438
+ "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 guy wearing a blue shirt and sunglasses appears behind the man in the black hat.,\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"
439
+ },
440
+ {
441
+ "from": "gpt",
442
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
443
+ }
444
+ ]
445
+ },
446
+ {
447
+ "id": "r104183",
448
+ "images": [
449
+ "r104183_00.jpg",
450
+ "r104183_01.jpg",
451
+ "r104183_02.jpg",
452
+ "r104183_03.jpg",
453
+ "r104183_04.jpg",
454
+ "r104183_05.jpg",
455
+ "r104183_06.jpg",
456
+ "r104183_07.jpg",
457
+ "r104183_08.jpg",
458
+ "r104183_09.jpg",
459
+ "r104183_10.jpg",
460
+ "r104183_11.jpg",
461
+ "r104183_12.jpg",
462
+ "r104183_13.jpg",
463
+ "r104183_14.jpg",
464
+ "r104183_15.jpg"
465
+ ],
466
+ "conversations": [
467
+ {
468
+ "from": "human",
469
+ "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 woman is sitting at a table in her bedroom and talking to the camera while holding a notebook and pen.,\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"
470
+ },
471
+ {
472
+ "from": "gpt",
473
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
474
+ }
475
+ ]
476
+ },
477
+ {
478
+ "id": "r102179",
479
+ "images": [
480
+ "r102179_00.jpg",
481
+ "r102179_01.jpg",
482
+ "r102179_02.jpg",
483
+ "r102179_03.jpg",
484
+ "r102179_04.jpg",
485
+ "r102179_05.jpg",
486
+ "r102179_06.jpg",
487
+ "r102179_07.jpg",
488
+ "r102179_08.jpg",
489
+ "r102179_09.jpg",
490
+ "r102179_10.jpg",
491
+ "r102179_11.jpg",
492
+ "r102179_12.jpg",
493
+ "r102179_13.jpg",
494
+ "r102179_14.jpg",
495
+ "r102179_15.jpg"
496
+ ],
497
+ "conversations": [
498
+ {
499
+ "from": "human",
500
+ "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 blue buckets next to each other on the ground.,\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"
501
+ },
502
+ {
503
+ "from": "gpt",
504
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
505
+ }
506
+ ]
507
+ },
508
+ {
509
+ "id": "r002473",
510
+ "images": [
511
+ "r002473_00.jpg",
512
+ "r002473_01.jpg",
513
+ "r002473_02.jpg",
514
+ "r002473_03.jpg",
515
+ "r002473_04.jpg",
516
+ "r002473_05.jpg",
517
+ "r002473_06.jpg",
518
+ "r002473_07.jpg",
519
+ "r002473_08.jpg",
520
+ "r002473_09.jpg",
521
+ "r002473_10.jpg",
522
+ "r002473_11.jpg",
523
+ "r002473_12.jpg",
524
+ "r002473_13.jpg",
525
+ "r002473_14.jpg",
526
+ "r002473_15.jpg",
527
+ "r002473_16.jpg",
528
+ "r002473_17.jpg",
529
+ "r002473_18.jpg",
530
+ "r002473_19.jpg",
531
+ "r002473_20.jpg",
532
+ "r002473_21.jpg",
533
+ "r002473_22.jpg",
534
+ "r002473_23.jpg"
535
+ ],
536
+ "conversations": [
537
+ {
538
+ "from": "human",
539
+ "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 no text is visible briefly.,\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"
540
+ },
541
+ {
542
+ "from": "gpt",
543
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
544
+ }
545
+ ]
546
+ },
547
+ {
548
+ "id": "r101103",
549
+ "images": [
550
+ "r101103_00.jpg",
551
+ "r101103_01.jpg",
552
+ "r101103_02.jpg",
553
+ "r101103_03.jpg",
554
+ "r101103_04.jpg",
555
+ "r101103_05.jpg",
556
+ "r101103_06.jpg",
557
+ "r101103_07.jpg",
558
+ "r101103_08.jpg",
559
+ "r101103_09.jpg",
560
+ "r101103_10.jpg",
561
+ "r101103_11.jpg",
562
+ "r101103_12.jpg",
563
+ "r101103_13.jpg",
564
+ "r101103_14.jpg",
565
+ "r101103_15.jpg",
566
+ "r101103_16.jpg",
567
+ "r101103_17.jpg",
568
+ "r101103_18.jpg",
569
+ "r101103_19.jpg",
570
+ "r101103_20.jpg",
571
+ "r101103_21.jpg",
572
+ "r101103_22.jpg",
573
+ "r101103_23.jpg"
574
+ ],
575
+ "conversations": [
576
+ {
577
+ "from": "human",
578
+ "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 person is playing a video game and mentions a gold player and a pack.,\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"
579
+ },
580
+ {
581
+ "from": "gpt",
582
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
583
+ }
584
+ ]
585
+ },
586
+ {
587
+ "id": "r100499",
588
+ "images": [
589
+ "r100499_00.jpg",
590
+ "r100499_01.jpg",
591
+ "r100499_02.jpg",
592
+ "r100499_03.jpg",
593
+ "r100499_04.jpg",
594
+ "r100499_05.jpg",
595
+ "r100499_06.jpg",
596
+ "r100499_07.jpg",
597
+ "r100499_08.jpg",
598
+ "r100499_09.jpg",
599
+ "r100499_10.jpg",
600
+ "r100499_11.jpg",
601
+ "r100499_12.jpg",
602
+ "r100499_13.jpg",
603
+ "r100499_14.jpg",
604
+ "r100499_15.jpg",
605
+ "r100499_16.jpg",
606
+ "r100499_17.jpg",
607
+ "r100499_18.jpg",
608
+ "r100499_19.jpg",
609
+ "r100499_20.jpg",
610
+ "r100499_21.jpg",
611
+ "r100499_22.jpg",
612
+ "r100499_23.jpg"
613
+ ],
614
+ "conversations": [
615
+ {
616
+ "from": "human",
617
+ "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 is a car driving down a long tunnel with bright headlights on.,\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"
618
+ },
619
+ {
620
+ "from": "gpt",
621
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
622
+ }
623
+ ]
624
+ },
625
+ {
626
+ "id": "r002810",
627
+ "images": [
628
+ "r002810_00.jpg",
629
+ "r002810_01.jpg",
630
+ "r002810_02.jpg",
631
+ "r002810_03.jpg",
632
+ "r002810_04.jpg",
633
+ "r002810_05.jpg",
634
+ "r002810_06.jpg",
635
+ "r002810_07.jpg",
636
+ "r002810_08.jpg",
637
+ "r002810_09.jpg",
638
+ "r002810_10.jpg",
639
+ "r002810_11.jpg",
640
+ "r002810_12.jpg",
641
+ "r002810_13.jpg",
642
+ "r002810_14.jpg",
643
+ "r002810_15.jpg",
644
+ "r002810_16.jpg",
645
+ "r002810_17.jpg",
646
+ "r002810_18.jpg",
647
+ "r002810_19.jpg",
648
+ "r002810_20.jpg",
649
+ "r002810_21.jpg",
650
+ "r002810_22.jpg",
651
+ "r002810_23.jpg"
652
+ ],
653
+ "conversations": [
654
+ {
655
+ "from": "human",
656
+ "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 scratches the dog's side.,\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"
657
+ },
658
+ {
659
+ "from": "gpt",
660
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
661
+ }
662
+ ]
663
+ },
664
+ {
665
+ "id": "r003265",
666
+ "images": [
667
+ "r003265_00.jpg",
668
+ "r003265_01.jpg",
669
+ "r003265_02.jpg",
670
+ "r003265_03.jpg",
671
+ "r003265_04.jpg",
672
+ "r003265_05.jpg",
673
+ "r003265_06.jpg",
674
+ "r003265_07.jpg",
675
+ "r003265_08.jpg",
676
+ "r003265_09.jpg",
677
+ "r003265_10.jpg",
678
+ "r003265_11.jpg",
679
+ "r003265_12.jpg",
680
+ "r003265_13.jpg",
681
+ "r003265_14.jpg",
682
+ "r003265_15.jpg",
683
+ "r003265_16.jpg",
684
+ "r003265_17.jpg",
685
+ "r003265_18.jpg",
686
+ "r003265_19.jpg",
687
+ "r003265_20.jpg",
688
+ "r003265_21.jpg",
689
+ "r003265_22.jpg",
690
+ "r003265_23.jpg"
691
+ ],
692
+ "conversations": [
693
+ {
694
+ "from": "human",
695
+ "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 one green bird flies 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"
696
+ },
697
+ {
698
+ "from": "gpt",
699
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
700
+ }
701
+ ]
702
+ },
703
+ {
704
+ "id": "r001443",
705
+ "images": [
706
+ "r001443_00.jpg",
707
+ "r001443_01.jpg",
708
+ "r001443_02.jpg",
709
+ "r001443_03.jpg",
710
+ "r001443_04.jpg",
711
+ "r001443_05.jpg",
712
+ "r001443_06.jpg",
713
+ "r001443_07.jpg",
714
+ "r001443_08.jpg",
715
+ "r001443_09.jpg",
716
+ "r001443_10.jpg",
717
+ "r001443_11.jpg",
718
+ "r001443_12.jpg",
719
+ "r001443_13.jpg",
720
+ "r001443_14.jpg",
721
+ "r001443_15.jpg",
722
+ "r001443_16.jpg",
723
+ "r001443_17.jpg",
724
+ "r001443_18.jpg",
725
+ "r001443_19.jpg",
726
+ "r001443_20.jpg",
727
+ "r001443_21.jpg",
728
+ "r001443_22.jpg",
729
+ "r001443_23.jpg"
730
+ ],
731
+ "conversations": [
732
+ {
733
+ "from": "human",
734
+ "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 Woman makes handmotions with baby.,\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"
735
+ },
736
+ {
737
+ "from": "gpt",
738
+ "value": "visual quality: 3\n object consistency: 3\n dynamic degree: 3\n motion smoothness: 3\n text-to-video alignment: 3\n factual consistency: 3\n overall score: 3\n"
739
+ }
740
+ ]
741
+ }
742
+ ]