task
stringclasses 1
value | source
stringclasses 1
value | video
stringlengths 37
38
| duration
int64 120
150
| tgt
sequencelengths 1
24
| conversations
listlengths 2
2
|
|---|---|---|---|---|---|
tvg | qvhighlights | qvh/videos/j7rJstUseKg_360.0_510.0.mp4 | 150 | [
[
72,
82
],
[
84,
94
],
[
96,
106
],
[
108,
118
],
[
120,
130
],
[
136,
142
],
[
144,
146
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'some military patriots takes us through their safety procedures and measures', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/j7rJstUseKg_210.0_360.0.mp4 | 150 | [
[
96,
114
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Man in baseball cap eats before doing his interview', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/j7rJstUseKg_60.0_210.0.mp4 | 150 | [
[
48,
50
],
[
76,
120
],
[
122,
138
],
[
140,
146
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A man in a white shirt discusses the right to have and carry firearms', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/-Oc6gSWB_HA_60.0_210.0.mp4 | 150 | [
[
64,
92
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A view of a bamboo fountain of water in a tea house and people scoop from and wash off' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/G60-kHBEeZA_60.0_210.0.mp4 | 150 | [
[
14,
48
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'The weather map shows large snowfall in the weather patterns' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/Ok-M_V_h-eY_210.0_360.0.mp4 | 150 | [
[
120,
134
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Kids are hanging off of a pole' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/Ok-M_V_h-eY_360.0_510.0.mp4 | 150 | [
[
14,
40
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'The view from the car window as it drives down the street' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/Ok-M_V_h-eY_60.0_210.0.mp4 | 150 | [
[
120,
134
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A woman is approaching an illuminate entrance with stairs at night' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S73Z-nM0GQE_60.0_210.0.mp4 | 150 | [
[
112,
126
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Woman holds up a small blue bottle' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S73Z-nM0GQE_210.0_360.0.mp4 | 150 | [
[
18,
32
],
[
60,
64
],
[
84,
98
],
[
126,
140
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A girl drinking a coffee in the kitchen' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S73Z-nM0GQE_360.0_510.0.mp4 | 150 | [
[
32,
78
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Egg is in a metal circle mold' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S73Z-nM0GQE_660.0_810.0.mp4 | 150 | [
[
90,
92
],
[
98,
114
],
[
122,
132
],
[
134,
140
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Woman holds up her white pants' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S73Z-nM0GQE_510.0_660.0.mp4 | 150 | [
[
0,
26
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'A lady with white top is holding a camera and talking in the room' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/tsOkWgzgW-o_60.0_210.0.mp4 | 150 | [
[
110,
114
],
[
124,
132
],
[
136,
142
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Bodies are lying in the rubble'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/tsOkWgzgW-o_210.0_360.0.mp4 | 150 | [
[
12,
26
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A man in military gear is posted up at an entrance while talking to a civilian' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/C0MQLh8Az7U_60.0_210.0.mp4 | 150 | [
[
6,
40
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Woman sits in the drivers seat of her car' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/C0MQLh8Az7U_360.0_510.0.mp4 | 150 | [
[
30,
56
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Kids pours liquid from a purple box into glass then tries it herself' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/C0MQLh8Az7U_210.0_360.0.mp4 | 150 | [
[
84,
120
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A young woman prepares some avocado toast', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S-nHYzK-BVg_360.0_510.0.mp4 | 150 | [
[
130,
142
],
[
148,
150
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'A skeleton ociture is in a word document' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S-nHYzK-BVg_60.0_210.0.mp4 | 150 | [
[
2,
16
],
[
106,
112
],
[
136,
140
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Different menus are shown on a computer screen'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S-nHYzK-BVg_210.0_360.0.mp4 | 150 | [
[
0,
150
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Computer screen is showing a word doc' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S-nHYzK-BVg_510.0_660.0.mp4 | 150 | [
[
0,
76
],
[
80,
150
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Picture of skeleton is on a word document' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/S-nHYzK-BVg_660.0_810.0.mp4 | 150 | [
[
0,
24
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'A demonstration of how to use photo software using an image of a skeleton' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/W8V0z-_dadg_360.0_510.0.mp4 | 150 | [
[
0,
24
],
[
28,
32
],
[
40,
48
],
[
52,
60
],
[
70,
84
],
[
104,
110
],
[
120,
126
],
[
130,
134
],
[
136,
144
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Man walks everywhere in a black t shirt', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/W8V0z-_dadg_210.0_360.0.mp4 | 150 | [
[
100,
116
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Man sets up his workspace then sits and works'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/W8V0z-_dadg_60.0_210.0.mp4 | 150 | [
[
96,
122
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'People are working out at the gym', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/LoLqz33MJNU_360.0_510.0.mp4 | 150 | [
[
2,
8
],
[
22,
28
],
[
32,
40
],
[
44,
50
],
[
56,
62
],
[
78,
84
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Asian chef with dyed pink hair cooks food' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/LoLqz33MJNU_210.0_360.0.mp4 | 150 | [
[
44,
56
],
[
60,
78
],
[
88,
136
],
[
146,
148
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Two asian chefs cooking a meal'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/LoLqz33MJNU_60.0_210.0.mp4 | 150 | [
[
0,
150
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A boy is taking cooking instructions from the super chef'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/LoLqz33MJNU_510.0_660.0.mp4 | 150 | [
[
136,
150
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Man in black top walks to the table then sits down' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/LoLqz33MJNU_660.0_810.0.mp4 | 150 | [
[
0,
118
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Two men share a bowl of noodles', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/bJZ-FTeG7D8_60.0_210.0.mp4 | 150 | [
[
0,
150
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Man is wearing a headset while interviewed to the Majority Report' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/bJZ-FTeG7D8_210.0_360.0.mp4 | 150 | [
[
0,
150
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Two people from the same show interview a man at his house' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/bJZ-FTeG7D8_360.0_510.0.mp4 | 150 | [
[
0,
150
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Man is wearing a headset during a interview with The Majority Report' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/WmId2ZP3h0c_360.0_510.0.mp4 | 150 | [
[
4,
20
],
[
28,
30
],
[
32,
46
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Man and woman having interview about indoctrination camp' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/WmId2ZP3h0c_60.0_210.0.mp4 | 150 | [
[
18,
38
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'A group of girls and boys are dancing' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/WmId2ZP3h0c_210.0_360.0.mp4 | 150 | [
[
98,
114
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Students wearing gender colored red and blue uniforms are attending a class' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/hu5L0-CnuUw_360.0_510.0.mp4 | 150 | [
[
36,
68
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A spoon stirs the yellow curry in a pot'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/hu5L0-CnuUw_210.0_360.0.mp4 | 150 | [
[
122,
150
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A little girl showing her hair and talking to her mom', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/hu5L0-CnuUw_60.0_210.0.mp4 | 150 | [
[
50,
56
],
[
66,
76
],
[
102,
112
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Mom does her daughter's makeup'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/hu5L0-CnuUw_660.0_810.0.mp4 | 150 | [
[
52,
76
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Little girl rides in the car with her mother', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/hu5L0-CnuUw_510.0_660.0.mp4 | 150 | [
[
38,
110
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A woman in green blouse is talking in front of the camera' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/q8YvpLSXQnk_360.0_510.0.mp4 | 150 | [
[
108,
150
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A lady capturing various attractive interior designs' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/q8YvpLSXQnk_60.0_210.0.mp4 | 150 | [
[
0,
24
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Two young women describe the sources of their bathing suits, cover ups, and hair' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/q8YvpLSXQnk_210.0_360.0.mp4 | 150 | [
[
0,
18
],
[
30,
34
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Two women in bathing suits walk along the beach together', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/qtyBzFV9yTs_210.0_360.0.mp4 | 150 | [
[
0,
70
],
[
72,
150
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A musical commercial on how to make the world a better place for us' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/qtyBzFV9yTs_60.0_210.0.mp4 | 150 | [
[
0,
30
],
[
78,
110
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Animation of the planet earth with some letters appearing on it' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/RZrf7QFxIW4_60.0_210.0.mp4 | 150 | [
[
94,
108
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Woman writes in a notebook' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/RZrf7QFxIW4_210.0_360.0.mp4 | 150 | [
[
126,
136
],
[
144,
150
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Masked woman walks around with her hood up' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/RZrf7QFxIW4_360.0_510.0.mp4 | 150 | [
[
12,
26
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A computer screen with text is shown' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/yGV2J8Fo2WQ_60.0_210.0.mp4 | 150 | [
[
24,
44
],
[
46,
88
],
[
90,
96
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Mother play with there children' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/yGV2J8Fo2WQ_360.0_510.0.mp4 | 150 | [
[
108,
116
],
[
118,
140
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A mother is bathing her two children in the bathtub', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/yGV2J8Fo2WQ_210.0_360.0.mp4 | 150 | [
[
28,
42
],
[
52,
82
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query '2 little girls home schooled by their mother' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/yGV2J8Fo2WQ_510.0_660.0.mp4 | 150 | [
[
18,
32
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A mother and her child are browsing through a library' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/yGV2J8Fo2WQ_660.0_810.0.mp4 | 150 | [
[
78,
100
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A lady is trying to make a dessert that looks just like the picture' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/9kwlibPTcJU_210.0_360.0.mp4 | 150 | [
[
66,
82
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Food is in tupperware and on plates' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/9kwlibPTcJU_360.0_510.0.mp4 | 150 | [
[
0,
72
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A girl is having fun with her friends' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/9kwlibPTcJU_60.0_210.0.mp4 | 150 | [
[
46,
50
],
[
52,
56
],
[
58,
60
],
[
64,
68
],
[
70,
72
],
[
76,
80
],
[
84,
86
],
[
90,
92
],
[
96,
102
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Blonde woman brushes her face' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/9kwlibPTcJU_510.0_660.0.mp4 | 150 | [
[
0,
28
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Two female classmates are joking around while in class together' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/9kwlibPTcJU_660.0_810.0.mp4 | 150 | [
[
2,
24
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'The lady with white shirt on is trying to put the ball point pen back together but couldn't' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/Mb_8DJF6Hp0_60.0_210.0.mp4 | 150 | [
[
2,
22
],
[
24,
120
],
[
126,
144
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Two men in hard hats have a conversation at a worksite' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/lAigvAeOoqQ_60.0_210.0.mp4 | 150 | [
[
0,
150
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Gil trying out different outfits at shop', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/lAigvAeOoqQ_360.0_510.0.mp4 | 150 | [
[
0,
144
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A girl talking while getting her make up' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/lAigvAeOoqQ_210.0_360.0.mp4 | 150 | [
[
78,
130
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Camilla Cabello introduces her hair and makeup people 2 hours before the VMAs', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/Kn1iFDv9Viw_60.0_210.0.mp4 | 150 | [
[
52,
54
],
[
120,
130
],
[
132,
140
],
[
144,
148
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A child lays on their back as a woman plays with them' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/JcHK1SmwDds_360.0_510.0.mp4 | 150 | [
[
52,
58
],
[
62,
78
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Woman is trying to cool off a man in a yelow shirt' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/JcHK1SmwDds_60.0_210.0.mp4 | 150 | [
[
0,
64
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Youtuber talking about his new tv commercial project', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/JcHK1SmwDds_210.0_360.0.mp4 | 150 | [
[
56,
150
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Man in yellow top films his medical appointment', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/JcHK1SmwDds_510.0_660.0.mp4 | 150 | [
[
70,
92
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A man is going through and tearing away pages on a tearaway calendar which is green' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/QndZGvTthvY_210.0_360.0.mp4 | 150 | [
[
70,
84
],
[
126,
146
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Dog is on a leash while interacting with other dogs', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/QndZGvTthvY_60.0_210.0.mp4 | 150 | [
[
74,
116
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A woman preparing an avocado sandwich' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/QndZGvTthvY_360.0_510.0.mp4 | 150 | [
[
62,
76
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Woman with towel on her head has on a red mask'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/QndZGvTthvY_660.0_810.0.mp4 | 150 | [
[
8,
22
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Woman picks up a jar of JIF'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/QndZGvTthvY_510.0_660.0.mp4 | 150 | [
[
122,
150
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A woman with a scarf is walking outside in the cold', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/BW0JOiuVt2o_210.0_360.0.mp4 | 150 | [
[
6,
36
],
[
64,
72
],
[
78,
102
],
[
106,
120
],
[
122,
124
],
[
126,
144
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A woman is bent over cleaning a floor', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/BW0JOiuVt2o_60.0_210.0.mp4 | 150 | [
[
0,
80
],
[
82,
124
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A woman cleans the floor of a wet-floor bathroom using a straw broom and water from an in-floor water tank' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/OKYptIe8a-k_360.0_510.0.mp4 | 150 | [
[
56,
82
],
[
96,
108
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A close up of the pot with the food being stirred' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/OKYptIe8a-k_60.0_210.0.mp4 | 150 | [
[
14,
40
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'A child is cleaning up and organizing his toys near his mother' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/OKYptIe8a-k_210.0_360.0.mp4 | 150 | [
[
24,
56
],
[
118,
130
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A woman peels and chops a banana for her son' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/OKYptIe8a-k_660.0_810.0.mp4 | 150 | [
[
50,
64
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Woman takes off her glasses and puts them back on' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/OKYptIe8a-k_510.0_660.0.mp4 | 150 | [
[
0,
34
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A mom and jer son eat as they watch TV together under the breeze of a ceiling fan', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/1z2wDjkPLq0_210.0_360.0.mp4 | 150 | [
[
112,
150
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A man wearing a turquoise t-shirt eating with his daughter while his son does homework at the table' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/1z2wDjkPLq0_60.0_210.0.mp4 | 150 | [
[
98,
150
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Two elementary school kids talking and eating some snacks in a car ', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/xrT84MJxBhs_210.0_360.0.mp4 | 150 | [
[
86,
110
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Phone screen showing an ad' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/xrT84MJxBhs_360.0_510.0.mp4 | 150 | [
[
104,
126
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'The girl is sitting cross legged on a chair with a cup of tea', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/xrT84MJxBhs_60.0_210.0.mp4 | 150 | [
[
0,
26
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A girl eating her breakfast with tea'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/xrT84MJxBhs_510.0_660.0.mp4 | 150 | [
[
64,
80
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A woman vlogger with white top is showing a tiny white dust bin'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/xrT84MJxBhs_660.0_810.0.mp4 | 150 | [
[
0,
124
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A woman in white t-shirt is talking about products in front of the camera'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/O16JP0YRKv4_210.0_360.0.mp4 | 150 | [
[
78,
98
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Woman interviews soldiers under a tent', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/O16JP0YRKv4_360.0_510.0.mp4 | 150 | [
[
50,
82
],
[
86,
96
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'Woman talks into her blue phone' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/O16JP0YRKv4_60.0_210.0.mp4 | 150 | [
[
34,
38
],
[
44,
78
]
] | [
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Soldier is prone looking through his rifle scope' and determine its start and end seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/0Cen89PVfhE_210.0_360.0.mp4 | 150 | [
[
48,
90
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A woman in red t-shirt and red hat is talking with a man in black t-shirt in front of a camera' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/0Cen89PVfhE_360.0_510.0.mp4 | 150 | [
[
54,
74
],
[
102,
150
]
] | [
{
"from": "human",
"value": "<image>\nThe given natural language query 'A man is being interviewed in front of a set of encyclopedia' is semantically aligned with a video moment, please give the start time and end time of the video moment.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/0Cen89PVfhE_60.0_210.0.mp4 | 150 | [
[
92,
112
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'President Trump is giving a speech at a conference behind a podium with a seal on it' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/8ySfrR3YQHM_60.0_210.0.mp4 | 150 | [
[
8,
150
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Group of school girls talking about compulsory mask' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/8ySfrR3YQHM_210.0_360.0.mp4 | 150 | [
[
118,
134
]
] | [
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Vlogger is looking at websites', when does the described content occur in the video? Please return the timestamp in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/8ySfrR3YQHM_360.0_510.0.mp4 | 150 | [
[
0,
14
],
[
22,
24
],
[
48,
62
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Man with brown sweatshirt is wearing a mask' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/8ySfrR3YQHM_510.0_660.0.mp4 | 150 | [
[
46,
64
]
] | [
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Girl looks at the water fountain before decided to drink from it' in the video, and output the start and end timestamps in seconds.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/8ySfrR3YQHM_660.0_810.0.mp4 | 150 | [
[
56,
70
]
] | [
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Girl is standing by a white board with drawings on it'.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
tvg | qvhighlights | qvh/videos/sJ-KomL7DUo_60.0_210.0.mp4 | 150 | [
[
96,
102
],
[
106,
150
]
] | [
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A girl enjoying fun water rides' within the video using timestamps.The format of your response should be: 'The event happens in <start time> - <end time>''."
},
{
"from": "gpt",
"value": "The event happens in <vid> - <vid>."
}
] |
End of preview. Expand
in Data Studio
No dataset card yet
- Downloads last month
- 7