dk-vlm / examples.json
multimodalart's picture
multimodalart HF Staff
Upload folder using huggingface_hub
5059823 verified
Raw History Blame Contribute Delete
4.95 kB
[
{
"video": "demos/vsi-d755b3d9d8.mp4",
"kind": "vsi",
"type": "route_planning",
"question": "The robot starts at the blue chair, facing the nearest monitor, and needs to reach the brown door. It turns and moves to the cabinet, then turns again and moves to the door. Which two turns should it make, in order?",
"options": [
"A. Turn Left, Turn Left",
"B. Turn Right, Turn Right",
"C. Turn Back, Turn Left",
"D. Turn Back, Turn Right"
],
"reference_answer": "D. Turn Back, Turn Right"
},
{
"video": "demos/vsi-d755b3d9d8.mp4",
"kind": "vsi",
"type": "object_rel_direction_hard",
"question": "Stand beside the door and face the cup. From your viewpoint, is the laptop front-left, front-right, back-left, or back-right?",
"options": [
"A. front-right",
"B. front-left",
"C. back-left",
"D. back-right"
],
"reference_answer": "B. front-left"
},
{
"video": "demos/arkitscenes-41069025.mp4",
"kind": "vsi",
"type": "object_counting",
"question": "How many chair(s) are in this room?",
"options": [],
"reference_answer": "2"
},
{
"video": "demos/vsi-3db0a1c8f3.mp4",
"kind": "vsi",
"type": "route_planning",
"question": "The robot starts at the toilet, facing the washing machine, and needs to reach the pan. It moves to the washing machine, turns toward the sofa, then turns again toward the pan. Which two turns should it make, in order?",
"options": [
"A. Turn Left, Turn Left",
"B. Turn Left, Turn Right",
"C. Turn Back, Turn Right",
"D. Turn Right, Turn Right"
],
"reference_answer": "D. Turn Right, Turn Right"
},
{
"video": "demos/vsi-3db0a1c8f3.mp4",
"kind": "vsi",
"type": "object_rel_distance",
"question": "Which is closest to the pan: the door, basket, table, or bed? Compare the shortest distance between the objects.",
"options": [
"A. door",
"B. basket",
"C. table",
"D. bed"
],
"reference_answer": "B. basket"
},
{
"video": "demos/vsi-5eb31827b7.mp4",
"kind": "vsi",
"type": "object_counting",
"question": "How many tables are in the room shown in the video?",
"options": [],
"reference_answer": "2"
},
{
"video": "demos/vsi-5eb31827b7.mp4",
"kind": "vsi",
"type": "object_size_estimation",
"question": "Estimate the door's longest dimension\u2014length, width, or height\u2014in centimeters.",
"options": [],
"reference_answer": "213 cm"
},
{
"video": "demos/vsi-scene0699_00.mp4",
"kind": "vsi",
"type": "obj_appearance_order",
"question": "Put the table, closet, door, and bed in the order in which they first appear in the video.",
"options": [
"A. door, bed, closet, table",
"B. bed, door, table, closet",
"C. closet, table, bed, door",
"D. table, closet, door, bed"
],
"reference_answer": "B. bed, door, table, closet"
},
{
"video": "demos/vsi-scene0699_00.mp4",
"kind": "vsi",
"type": "object_abs_distance",
"question": "Approximately how many meters separate the closest points of the table and the bed?",
"options": [],
"reference_answer": "1.0 m"
},
{
"video": "demos/rc-b8d441d736.mp4",
"kind": "rc",
"type": "rc_composition_action",
"question": "start: tv\nfacing: door\nsequence: heater -> trash can\nOutput the turn at each stage as a JSON array.",
"options": [],
"reference_answer": "[\"left\", \"back\"]"
},
{
"video": "demos/rc-b8d441d736.mp4",
"kind": "rc",
"type": "rc_composition_angle",
"question": "start: door\nfacing: trash can\nsequence: tv -> trash can\nOutput the clockwise rotation in degrees at each stage, as a JSON array of integers.",
"options": [],
"reference_answer": "[14, 191]"
},
{
"video": "demos/rc-scene0055_00.mp4",
"kind": "rc",
"type": "rc_composition_action",
"question": "start: chair\nfacing: bed\nsequence: window -> door\nOutput the turn at each stage as a JSON array.",
"options": [],
"reference_answer": "[\"right\", \"left\"]"
},
{
"video": "demos/rc-scene0055_00.mp4",
"kind": "rc",
"type": "rc_composition_angle",
"question": "start: bed\nfacing: window\nsequence: chair -> door\nOutput the clockwise rotation in degrees at each stage, as a JSON array of integers.",
"options": [],
"reference_answer": "[23, 211]"
},
{
"video": "demos/rc-37ea1c52f0.mp4",
"kind": "rc",
"type": "rc_composition_action",
"question": "origin=telephone\ninitial_heading_anchor=plant\nordered_landmarks=[power strip, plant, tv]\nReturn the relative heading changes in order as a JSON array.",
"options": [],
"reference_answer": "[\"left\", \"back\", \"back\"]"
},
{
"video": "demos/rc-37ea1c52f0.mp4",
"kind": "rc",
"type": "rc_composition_angle",
"question": "Starting at telephone and initially oriented toward plant, visit these landmarks in order: power strip, plant, tv.\nGive the clockwise turn in degrees required before each landmark, as a JSON array of integers.",
"options": [],
"reference_answer": "[266, 166, 153]"
}
]