task
stringclasses 1
value | source
stringclasses 1
value | video
stringlengths 37
38
| duration
int64 120
150
| tgt
listlengths 1
24
| conversations
listlengths 2
2
|
|---|---|---|---|---|---|
tvg
|
qvhighlights
|
qvh/videos/vh0ZcfGjQSA_510.0_660.0.mp4
| 144 |
[
[
84,
136
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A woman in a black jacket sits in a hanging jacket in a room talking to a 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/5X-U3Yb5BqU_60.0_210.0.mp4
| 150 |
[
[
16,
18
],
[
48,
62
],
[
64,
110
],
[
112,
116
],
[
118,
150
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Woman vlogs while holding a drink with whipped cream'.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/5X-U3Yb5BqU_210.0_360.0.mp4
| 150 |
[
[
82,
84
],
[
90,
120
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A young woman showing her balcony through the glass door'.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/5X-U3Yb5BqU_360.0_510.0.mp4
| 150 |
[
[
4,
8
],
[
10,
14
],
[
106,
122
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Woman holds up a pink cream' 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/5X-U3Yb5BqU_660.0_810.0.mp4
| 150 |
[
[
126,
150
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A woman talks about an empty pink box of insanely yummy chocolate chip biscuits' 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/5X-U3Yb5BqU_510.0_660.0.mp4
| 150 |
[
[
136,
150
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A woman is going through a picture gallery on her phone', 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/9o3yvwD5KC8_60.0_210.0.mp4
| 150 |
[
[
122,
142
],
[
144,
150
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A lady's selfie video with her kid from a living room' 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/9o3yvwD5KC8_360.0_510.0.mp4
| 150 |
[
[
42,
92
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A woman shows her home workout using a 10 pound bag of sugar', 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/9o3yvwD5KC8_210.0_360.0.mp4
| 150 |
[
[
102,
118
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Video shows a clothes website' 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/9o3yvwD5KC8_660.0_810.0.mp4
| 150 |
[
[
74,
92
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Woman puts on a blue hairwrap' 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/9o3yvwD5KC8_510.0_660.0.mp4
| 150 |
[
[
24,
64
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Woman prepares the coffee maker'.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/moCYwUhOdts_60.0_210.0.mp4
| 150 |
[
[
44,
74
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Restaurant worker makes a desert' 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/moCYwUhOdts_210.0_360.0.mp4
| 150 |
[
[
114,
142
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Man and woman chill by the pool' 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/7U2R-gJmJqA_60.0_210.0.mp4
| 150 |
[
[
86,
132
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Mom is holding her son against her side', 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/7U2R-gJmJqA_360.0_510.0.mp4
| 150 |
[
[
46,
66
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Woman looks through clothes on a rack' 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/7U2R-gJmJqA_210.0_360.0.mp4
| 150 |
[
[
72,
84
],
[
88,
96
],
[
98,
150
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Blonde woman vlogs with large curtains in the background' 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/7U2R-gJmJqA_510.0_660.0.mp4
| 150 |
[
[
40,
42
],
[
58,
72
],
[
80,
86
]
] |
[
{
"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 a blouse shirt shows the baby she's holding on 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/7U2R-gJmJqA_660.0_810.0.mp4
| 150 |
[
[
10,
38
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A room is filled with storage and a crib' 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/2MJ9-APRapA_60.0_210.0.mp4
| 150 |
[
[
0,
50
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'woman talking to camera in passenger seat of caravan' 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/2MJ9-APRapA_360.0_510.0.mp4
| 150 |
[
[
24,
78
],
[
80,
82
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'There's a fireworks display over the water' 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/2MJ9-APRapA_210.0_360.0.mp4
| 150 |
[
[
0,
108
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Dash cam view of a bus during travel' 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/2MJ9-APRapA_510.0_660.0.mp4
| 150 |
[
[
24,
78
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Dogs are in a white pen', 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/wEbLiJ1A6vg_360.0_510.0.mp4
| 150 |
[
[
0,
150
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A lady talking about her favourite travel places' 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/wEbLiJ1A6vg_210.0_360.0.mp4
| 150 |
[
[
110,
148
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Travel vlogger \"sologirlstravelguide\" talks about Sri Lanka, a travel destination not yet ruined by tourism' 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/wEbLiJ1A6vg_60.0_210.0.mp4
| 150 |
[
[
26,
40
],
[
42,
44
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Woman in black top opens a beer and drinks it' 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/FFf9jfzSIzA_60.0_210.0.mp4
| 150 |
[
[
114,
136
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A girl talks about her travel bag which she love', 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/FFf9jfzSIzA_360.0_510.0.mp4
| 150 |
[
[
64,
70
],
[
76,
94
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Black woman has blonde highlights in her curls', 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/FFf9jfzSIzA_210.0_360.0.mp4
| 150 |
[
[
0,
44
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Girls having fun on breakfast'.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/FFf9jfzSIzA_510.0_660.0.mp4
| 150 |
[
[
104,
150
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A lady with white cardigan is sitting on the chair and talking about her experience' 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/FFf9jfzSIzA_660.0_810.0.mp4
| 150 |
[
[
66,
68
],
[
98,
110
],
[
118,
120
],
[
128,
130
],
[
132,
142
],
[
144,
146
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Blonde woman is having a conversation with a woman in beige sweater', 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/EOdbFy2nuhQ_60.0_210.0.mp4
| 150 |
[
[
26,
38
],
[
40,
46
],
[
48,
94
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Animations are showing different numbers in yellow'.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/EOdbFy2nuhQ_210.0_360.0.mp4
| 150 |
[
[
48,
64
],
[
108,
120
],
[
126,
128
],
[
134,
140
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Man in gray t shirt reports from his office', 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/EOdbFy2nuhQ_360.0_510.0.mp4
| 150 |
[
[
86,
114
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Children are skating on the sidewalk', 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/vzsFBzelssA_60.0_210.0.mp4
| 150 |
[
[
0,
60
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Chinese girls enjoying at some beautiful place'.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/vzsFBzelssA_210.0_360.0.mp4
| 150 |
[
[
122,
128
],
[
132,
150
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Woman in pink dress and a woman in a leather jacket pose for pictures in a stream' 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/vzsFBzelssA_360.0_510.0.mp4
| 150 |
[
[
34,
50
],
[
58,
68
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Woman wears a leather jacket in a moving truck bed', 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/vzsFBzelssA_660.0_810.0.mp4
| 150 |
[
[
136,
150
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A lady with long hair is pointing out the mouth ulcer on her lips', 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/vzsFBzelssA_510.0_660.0.mp4
| 150 |
[
[
76,
86
],
[
92,
110
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A guy with red top is cooking curry dish' 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/81bW2WoY-gU_360.0_510.0.mp4
| 150 |
[
[
106,
134
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A woman harvests the field' 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/81bW2WoY-gU_60.0_210.0.mp4
| 150 |
[
[
108,
150
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'A group of travellers go talk to a woman picking taros in a field' 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/81bW2WoY-gU_210.0_360.0.mp4
| 150 |
[
[
136,
150
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A woman with short hair is harvesting a type of plant in the field and explaining her process' 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/rqrFAfl5o0Q_210.0_360.0.mp4
| 150 |
[
[
10,
14
],
[
26,
32
],
[
46,
56
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A family playing and having fun on the beach' 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/rqrFAfl5o0Q_360.0_510.0.mp4
| 150 |
[
[
84,
94
],
[
96,
100
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Family travels a river by boat' 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/rqrFAfl5o0Q_60.0_210.0.mp4
| 150 |
[
[
114,
128
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Family is circling around and exercising together at the beach' 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/MZ1S9oi-zDw_360.0_510.0.mp4
| 150 |
[
[
30,
44
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'A couple of news anchors are reporting from a circular white desk' 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/MZ1S9oi-zDw_60.0_210.0.mp4
| 150 |
[
[
106,
110
],
[
112,
138
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'An infographic display of England's Coronavirus restrictions is discussed'.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/MZ1S9oi-zDw_210.0_360.0.mp4
| 150 |
[
[
98,
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 news anchor is giving a news report in front of an orange scene of a palace at sunset'.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/38dpubUrrJs_360.0_510.0.mp4
| 142 |
[
[
106,
134
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'Couple having fun at party' 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/38dpubUrrJs_60.0_210.0.mp4
| 150 |
[
[
0,
150
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'A couple talking in a car while the lady drives the car' 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/38dpubUrrJs_210.0_360.0.mp4
| 150 |
[
[
0,
50
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A woman drives a car while a man talks' 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/k-cX11Cul5s_360.0_510.0.mp4
| 150 |
[
[
28,
150
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A blonde woman with black jacket and red scarf is talking in front of a camera with a microphone in a grassy field' 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/HtPQ421LFxE_210.0_360.0.mp4
| 150 |
[
[
88,
102
],
[
112,
118
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A plate of cinnamon rolls is on the table' 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/HtPQ421LFxE_60.0_210.0.mp4
| 150 |
[
[
72,
82
],
[
88,
98
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A couple is shopping the meat section of a store'.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/HtPQ421LFxE_360.0_510.0.mp4
| 150 |
[
[
136,
150
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Man in back of restaurant is playing guitar', 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/-iW7ZRXn9UM_210.0_360.0.mp4
| 150 |
[
[
36,
70
],
[
74,
76
],
[
82,
94
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A cute toddler climbing on a staircase' 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/-iW7ZRXn9UM_60.0_210.0.mp4
| 150 |
[
[
40,
72
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A man in a ball cap and red shirt holds his toddler in the air as the toddler 'dances'' 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/-iW7ZRXn9UM_360.0_510.0.mp4
| 150 |
[
[
130,
150
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'A little boy uses a monkey face filter on the 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/jj65-UxZMsM_60.0_210.0.mp4
| 150 |
[
[
18,
66
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A man in a black top is putting on a scooter helmet in a back street, preparing to ride to a coffee shop'.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/jj65-UxZMsM_360.0_510.0.mp4
| 150 |
[
[
36,
130
],
[
132,
134
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Shirtless man talks while standing in a pool' 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/jj65-UxZMsM_210.0_360.0.mp4
| 150 |
[
[
12,
90
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A happy couple talk about the wonderful breakfast smells as they sit at a busy cafe' 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/WWlUFdZ9Otk_360.0_510.0.mp4
| 150 |
[
[
60,
108
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A black suitcase is opened and it's orange interior compartments are shown' 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/WWlUFdZ9Otk_60.0_210.0.mp4
| 150 |
[
[
134,
150
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A woman is referencing a phone while writing on a paper on a desk' 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/WWlUFdZ9Otk_210.0_360.0.mp4
| 150 |
[
[
0,
24
],
[
36,
150
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Woman shows different pieces of paper'.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/3972B7evdK0_360.0_510.0.mp4
| 150 |
[
[
66,
84
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Pop stars dance in a concert' 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/3972B7evdK0_60.0_210.0.mp4
| 150 |
[
[
104,
144
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A group of friends open and try some saltwater taffy' 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/3972B7evdK0_210.0_360.0.mp4
| 150 |
[
[
42,
70
],
[
72,
74
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'People get bags of M&Ms from the M&M store,' 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/Jd0FrNZ4BUE_60.0_210.0.mp4
| 150 |
[
[
0,
150
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Different text is shown to the left of three headshots' 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/Jd0FrNZ4BUE_210.0_360.0.mp4
| 136 |
[
[
0,
132
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'Lots of text with three men's headshots to the right'.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/IzouSaF80Bw_210.0_360.0.mp4
| 150 |
[
[
0,
34
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A family is spending time at an easter festival' 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/IzouSaF80Bw_360.0_510.0.mp4
| 150 |
[
[
126,
140
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'As a child shakes her hands like \"claws\" as she watches a float go by on the street', 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/IzouSaF80Bw_60.0_210.0.mp4
| 150 |
[
[
64,
96
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'The children pout as a family lines up to enter Hong Kong Disneyland in the rain', 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/IzouSaF80Bw_660.0_810.0.mp4
| 150 |
[
[
84,
92
],
[
98,
110
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'a neon float with circling sharks goes by', 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/ICLPyhSdPlA_60.0_210.0.mp4
| 148 |
[
[
0,
12
],
[
14,
48
],
[
64,
84
],
[
88,
98
],
[
102,
116
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A video showing a forest on fire' 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/7OfS3ULVmg8_60.0_210.0.mp4
| 150 |
[
[
100,
114
],
[
124,
130
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Woman washes her dishes in the sink' 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/7OfS3ULVmg8_360.0_510.0.mp4
| 150 |
[
[
18,
52
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A group of friends are eating dinner 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/7OfS3ULVmg8_210.0_360.0.mp4
| 150 |
[
[
10,
28
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Woman talks holding a white coffee mug' 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/I0DX-dCN9DM_60.0_210.0.mp4
| 150 |
[
[
90,
128
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'The vlogger shows how to delete writing in the frame' 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/I0DX-dCN9DM_210.0_360.0.mp4
| 150 |
[
[
82,
86
],
[
88,
102
],
[
120,
150
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'Video editing software shows a road' 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/I0DX-dCN9DM_360.0_510.0.mp4
| 150 |
[
[
10,
28
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A keyboard application is used to input text' 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/_RWd0o7MoNQ_60.0_210.0.mp4
| 150 |
[
[
38,
44
],
[
50,
56
],
[
62,
84
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'A weather lady wearing a dress describes the winter storm watch and cold front that are arriving' 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/6fck0P4Rys8_60.0_210.0.mp4
| 150 |
[
[
52,
80
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Man with sunglasses waits outside', 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/6fck0P4Rys8_210.0_360.0.mp4
| 150 |
[
[
86,
98
],
[
108,
120
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A monkey is hanging out on the stairs and balcony of the bungalow'.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/6fck0P4Rys8_360.0_510.0.mp4
| 150 |
[
[
104,
108
],
[
110,
150
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Waves are coming in and out of the beach' 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/6fck0P4Rys8_510.0_660.0.mp4
| 150 |
[
[
6,
150
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Talk talks to the camera in a alleyway with a motorcycle' 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/DPQiB_x-V7s_360.0_510.0.mp4
| 150 |
[
[
40,
52
],
[
90,
92
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'Vlogger swims under the water and was on a boat', 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/DPQiB_x-V7s_210.0_360.0.mp4
| 150 |
[
[
0,
4
],
[
6,
20
],
[
32,
60
],
[
68,
120
],
[
126,
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 mam wears a bandana while sitting on the floor'.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/DPQiB_x-V7s_60.0_210.0.mp4
| 150 |
[
[
56,
58
],
[
60,
74
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Girl is hugging her dad from behind' 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/YKahVpkAk08_360.0_510.0.mp4
| 150 |
[
[
30,
32
],
[
36,
52
],
[
70,
72
],
[
74,
76
]
] |
[
{
"from": "human",
"value": "<image>\nDetect and report the start and end timestamps of the video segment that semantically matches the given textual query 'A group of men are hanging out in a room filled with blue LED lights'.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/YKahVpkAk08_60.0_210.0.mp4
| 150 |
[
[
6,
24
],
[
32,
36
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Two masked men escort a teen in a red sweatshirt into a house' 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/YKahVpkAk08_210.0_360.0.mp4
| 150 |
[
[
6,
10
],
[
12,
16
],
[
20,
22
],
[
30,
34
],
[
40,
42
],
[
46,
48
],
[
50,
52
],
[
58,
64
],
[
88,
90
],
[
94,
96
],
[
102,
104
],
[
124,
126
],
[
128,
132
],
[
140,
142
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'Comic book text is providing context for the friend's hanging out in the house' 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/WC9DPsSruAA_210.0_360.0.mp4
| 150 |
[
[
2,
84
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A lady explaining about the calm birth method for pregnant women', 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/WC9DPsSruAA_60.0_210.0.mp4
| 150 |
[
[
116,
132
],
[
138,
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 woman cooks food and gives a commentary'.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/WC9DPsSruAA_360.0_510.0.mp4
| 150 |
[
[
128,
136
],
[
140,
150
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'Dog wanders on the floor to get fed' 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/WC9DPsSruAA_660.0_810.0.mp4
| 150 |
[
[
34,
56
]
] |
[
{
"from": "human",
"value": "<image>\nGiven a textual query: 'A woman is showing a small note written in red ink on a tag', 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/WC9DPsSruAA_510.0_660.0.mp4
| 150 |
[
[
66,
150
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'A woman is showing her baby products' 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/OzZVM6obklA_360.0_510.0.mp4
| 150 |
[
[
60,
118
]
] |
[
{
"from": "human",
"value": "<image>\nThe given natural language query 'A man shows food that is inside of a box' 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/OzZVM6obklA_60.0_210.0.mp4
| 150 |
[
[
80,
120
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A man eats a Nile river fish dish outside the Bob Marley Guest house in Aswan' 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/OzZVM6obklA_210.0_360.0.mp4
| 150 |
[
[
36,
72
]
] |
[
{
"from": "human",
"value": "<image>\nFind the video segment that corresponds to the given textual query 'A man in blue t-shirt as well as wearing a black sunglasses is talking in front of a camera' 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/9bQ98HLDQ38_210.0_360.0.mp4
| 150 |
[
[
0,
26
]
] |
[
{
"from": "human",
"value": "<image>\nLocalize the visual content described by the given textual query 'A youtuber's selfie video with a bird and a monkey' 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/9bQ98HLDQ38_360.0_510.0.mp4
| 150 |
[
[
96,
150
]
] |
[
{
"from": "human",
"value": "<image>\nLocate the visual content mentioned in the text query 'A woman in a yellow mask is touring a museum with different artifacts and fossils on display' 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>."
}
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.