file_name stringlengths 31 31 | im_path stringlengths 46 46 | caption stringlengths 2 104 | bbox stringlengths 21 45 | prompt stringlengths 2.03k 2.52k | texts stringlengths 2 61 | sentence stringlengths 42 524 | outp_org listlengths 4 4 | formatted_sentence stringlengths 37 71 | modified_sentence stringlengths 38 518 | target listlengths 4 4 |
|---|---|---|---|---|---|---|---|---|---|---|
COCO_train2014_000000551717.jpg | refcoco_images/COCO_train2014_000000551717.jpg | ticket taker | [575.14,118.38,640.,,,381.08] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Ticket', 'Taker'] | Detected Ticket at location [347, 144, 495, 401]
Detected Taker at location [347, 144, 496, 401]
Detected ticket taker at location [575, 118, 639, 380]
| [
575,
118,
639,
380
] | Detected ticket taker at location [575, 118, 639, 380] | Detected Ticket at location [347, 144, 495, 401]
Detected Taker at location [347, 144, 496, 401]
Detected ticket taker at location [575, 118, 639, 380] | [
575,
118,
639,
380
] |
COCO_train2014_000000518957.jpg | refcoco_images/COCO_train2014_000000518957.jpg | furthest left person | [,31.74,,75.02,137.54,348.29] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Person'] | Detected Person at location [520, 32, 638, 391]
Detected Person at location [243, 123, 408, 383]
Detected furthest left person at location [32, 76, 135, 348]
| [
32,
76,
135,
348
] | Detected person at location [32, 76, 135, 348] | Detected Person at location [520, 32, 638, 391]
Detected Person at location [243, 123, 408, 383]
Detected person at location [32, 76, 135, 348] | [
32,
76,
135,
348
] |
COCO_train2014_000000547633.jpg | refcoco_images/COCO_train2014_000000547633.jpg | mug on right next to spoons | [316.42,293.65,475.35,433.96] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Mug', 'Spoon'] | Detected Mug at location [319, 292, 475, 433]
Detected Mug at location [292, 162, 430, 295]
Detected Spoon at location [331, 410, 436, 606]
Detected Spoon at location [313, 366, 359, 570]
Detected mug on right next to spoons at location [318, 294, 474, 433]
| [
318,
294,
474,
433
] | Detected mug at location [318, 294, 474, 433] | Detected Mug at location [319, 292, 475, 433]
Detected Mug at location [292, 162, 430, 295]
Detected Spoon at location [331, 410, 436, 606]
Detected Spoon at location [313, 366, 359, 570]
Detected mug at location [318, 294, 474, 433] | [
318,
294,
474,
433
] |
COCO_train2014_000000553038.jpg | refcoco_images/COCO_train2014_000000553038.jpg | Puffin on the right stretching | [408.5,,,,0.,,,563.51,161.39] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Puffin'] | Detected Puffin at location [191, 273, 364, 410]
Detected Puffin on the right stretching at location [409, 0, 562, 160]
| [
409,
0,
562,
160
] | Detected Puffin at location [409, 0, 562, 160] | Detected Puffin at location [191, 273, 364, 410]
Detected Puffin at location [409, 0, 562, 160] | [
409,
0,
562,
160
] |
COCO_train2014_000000566495.jpg | refcoco_images/COCO_train2014_000000566495.jpg | 1st horse | [,,0.,,,232.57,274.22,564.24] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Horse'] | Detected Horse at location [0, 233, 273, 638]
Detected Horse at location [0, 234, 273, 563]
Detected 1st horse at location [274, 261, 427, 572]
| [
274,
261,
427,
572
] | Detected horse at location [274, 261, 427, 572] | Detected Horse at location [0, 233, 273, 638]
Detected Horse at location [0, 234, 273, 563]
Detected horse at location [274, 261, 427, 572] | [
274,
261,
427,
572
] |
COCO_train2014_000000540947.jpg | refcoco_images/COCO_train2014_000000540947.jpg | hitter | [,70.74,225.05,256.5,,469.82] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Hitter'] | Detected Hitter at location [11, 221, 260, 470]
Detected Hitter at location [70, 222, 258, 469]
Detected hitter at location [71, 225, 255, 467]
| [
71,
225,
255,
467
] | Detected hitter at location [71, 225, 255, 467] | Detected Hitter at location [11, 221, 260, 470]
Detected Hitter at location [70, 222, 258, 469]
Detected hitter at location [71, 225, 255, 467] | [
71,
225,
255,
467
] |
COCO_train2014_000000547135.jpg | refcoco_images/COCO_train2014_000000547135.jpg | front most bike not the people just the bike | [,,0.94,317.27,347.39,418.,,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bike'] | Detected Bike at location [2, 273, 347, 416]
Detected Bike at location [306, 202, 473, 416]
Detected front most bike at location [306, 206, 469, 416]
Detected not the people at location [0, 309, 346, 416]
Detected just the bike at location [0, 171, 160, 362]
| [
0,
171,
160,
362
] | Detected bike at location [0, 171, 160, 362] | Detected Bike at location [2, 273, 347, 416]
Detected Bike at location [306, 202, 473, 416]
Detected front most bike at location [306, 206, 469, 416]
Detected not the people at location [0, 309, 346, 416]
Detected bike at location [0, 171, 160, 362] | [
0,
171,
160,
362
] |
COCO_train2014_000000539727.jpg | refcoco_images/COCO_train2014_000000539727.jpg | donut with colored sprinkles | [351.66,,75.38,568.95,289.81] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Donut', 'Sprinkles'] | Detected Donut at location [291, 0, 639, 423]
Detected Donut at location [354, 75, 569, 286]
Detected sprinkles at location [356, 77, 567, 285]
Detected donut with colored sprinkles at location [352, 76, 568, 290]
| [
352,
76,
568,
290
] | Detected donut at location [352, 76, 568, 290] | Detected Donut at location [291, 0, 639, 423]
Detected Donut at location [354, 75, 569, 286]
Detected sprinkles at location [356, 77, 567, 285]
Detected donut at location [352, 76, 568, 290] | [
352,
76,
568,
290
] |
COCO_train2014_000000534926.jpg | refcoco_images/COCO_train2014_000000534926.jpg | man on bikre | [198.96,,84.6,,306.8,,384.9,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Man', 'Bike'] | Detected Man at location [144, 78, 487, 392]
Detected Man at location [199, 80, 316, 386]
Detected Bike at location [144, 115, 486, 392]
Detected man on bikre at location [199, 84, 310, 384]
| [
199,
84,
310,
384
] | Detected man at location [199, 84, 310, 384] | Detected Man at location [144, 78, 487, 392]
Detected Man at location [199, 80, 316, 386]
Detected Bike at location [144, 115, 486, 392]
Detected man at location [199, 84, 310, 384] | [
199,
84,
310,
384
] |
COCO_train2014_000000527509.jpg | refcoco_images/COCO_train2014_000000527509.jpg | with green and red spot | [,49.32,,68.95,216.09,173.37] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Spot'] | Detected Spot at location [3, 3, 639, 479]
Detected with green and red spot at location [49, 68, 214, 173]
| [
49,
68,
214,
173
] | Detected green spot at location [49, 68, 214, 173] | Detected Spot at location [3, 3, 639, 479]
Detected green spot at location [49, 68, 214, 173] | [
49,
68,
214,
173
] |
COCO_train2014_000000572602.jpg | refcoco_images/COCO_train2014_000000572602.jpg | the bowl of strawberries | [,,0.,,,152.64,322.36,469.66] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bowl', 'Strawberries'] | Detected Bowl at location [0, 0, 575, 474]
Detected Bowl at location [0, 153, 325, 474]
Detected the bowl of strawberries at location [0, 154, 322, 474]
| [
0,
154,
322,
474
] | Detected bowl at location [0, 154, 322, 474] | Detected Bowl at location [0, 0, 575, 474]
Detected Bowl at location [0, 153, 325, 474]
Detected bowl at location [0, 154, 322, 474] | [
0,
154,
322,
474
] |
COCO_train2014_000000575711.jpg | refcoco_images/COCO_train2014_000000575711.jpg | woman in navy shirt with birds on it | [,,1.,,,,81.38,261.81,435.77] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Navy shirt', 'Birds'] | Detected Navy shirt at location [3, 202, 194, 440]
Detected woman in navy shirt with birds on it at location [0, 82, 262, 440]
| [
0,
82,
262,
440
] | Detected woman at location [0, 82, 262, 440] | Detected Navy shirt at location [3, 202, 194, 440]
Detected woman at location [0, 82, 262, 440] | [
0,
82,
262,
440
] |
COCO_train2014_000000562092.jpg | refcoco_images/COCO_train2014_000000562092.jpg | cow on right | [540.2,,166.69,640.,,,350.15] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Cow'] | Detected Cow at location [83, 13, 633, 475]
Detected cow on right at location [540, 167, 639, 351]
| [
540,
167,
639,
351
] | Detected cow at location [540, 167, 639, 351] | Detected Cow at location [83, 13, 633, 475]
Detected cow at location [540, 167, 639, 351] | [
540,
167,
639,
351
] |
COCO_train2014_000000541721.jpg | refcoco_images/COCO_train2014_000000541721.jpg | right person | [304.74,,51.62,523.17,420.33] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Person'] | Detected Person at location [302, 50, 522, 420]
Detected Person at location [105, 46, 317, 421]
Detected right person at location [304, 52, 521, 420]
| [
304,
52,
521,
420
] | Detected person at location [304, 52, 521, 420] | Detected Person at location [302, 50, 522, 420]
Detected Person at location [105, 46, 317, 421]
Detected person at location [304, 52, 521, 420] | [
304,
52,
521,
420
] |
COCO_train2014_000000566368.jpg | refcoco_images/COCO_train2014_000000566368.jpg | bottom middle | [152.81,251.5,,317.98,495.32] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | [] | Detected remote control at location [2, 1, 479, 639]
Detected bottom middle at location [2, 2, 479, 639]
| [
2,
2,
479,
639
] | Detected at location [2, 2, 479, 639] | Detected remote control at location [2, 1, 479, 639]
Detected at location [2, 2, 479, 639] | [
2,
2,
479,
639
] |
COCO_train2014_000000571644.jpg | refcoco_images/COCO_train2014_000000571644.jpg | mini couch kids are sitting on | [,51.76,189.94,402.34,369.63] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Couch', 'Kids'] | Detected Couch at location [67, 245, 518, 479]
Detected Kids at location [323, 106, 487, 293]
Detected Kids at location [107, 133, 223, 283]
Detected mini couch kids are sitting on at location [67, 244, 514, 479]
| [
67,
244,
514,
479
] | Detected mini couch at location [67, 244, 514, 479] | Detected Couch at location [67, 245, 518, 479]
Detected Kids at location [323, 106, 487, 293]
Detected Kids at location [107, 133, 223, 283]
Detected mini couch at location [67, 244, 514, 479] | [
67,
244,
514,
479
] |
COCO_train2014_000000533484.jpg | refcoco_images/COCO_train2014_000000533484.jpg | white nerd | [212.34,,69.16,375.97,368.18] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Nerd'] | Detected Nerd at location [268, 86, 494, 479]
Detected white nerd at location [271, 89, 480, 479]
| [
271,
89,
480,
479
] | Detected white nerd at location [271, 89, 480, 479] | Detected Nerd at location [268, 86, 494, 479]
Detected white nerd at location [271, 89, 480, 479] | [
271,
89,
480,
479
] |
COCO_train2014_000000520434.jpg | refcoco_images/COCO_train2014_000000520434.jpg | right gal | [421.32,,24.94,593.15,415.2,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Gal'] | Detected Gal at location [107, 22, 284, 411]
Detected Gal at location [420, 24, 593, 412]
Detected right gal at location [422, 26, 591, 411]
| [
422,
26,
591,
411
] | Detected right gal at location [422, 26, 591, 411] | Detected Gal at location [107, 22, 284, 411]
Detected Gal at location [420, 24, 593, 412]
Detected right gal at location [422, 26, 591, 411] | [
422,
26,
591,
411
] |
COCO_train2014_000000542159.jpg | refcoco_images/COCO_train2014_000000542159.jpg | left | [,,1.11,,49.03,241.8,,423.43] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | [] | Detected left at location [0, 50, 241, 427]
| [
0,
50,
241,
427
] | Detected at location [0, 50, 241, 427] | Detected at location [0, 50, 241, 427] | [
0,
50,
241,
427
] |
COCO_train2014_000000574211.jpg | refcoco_images/COCO_train2014_000000574211.jpg | guy on right | [451.68,103.49,563.17,369.35] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Guy'] | Detected Guy at location [448, 102, 566, 373]
Detected Guy at location [326, 107, 439, 353]
Detected Guy at location [219, 106, 327, 358]
Detected Guy at location [112, 101, 213, 367]
Detected guy on right at location [451, 103, 562, 370]
| [
451,
103,
562,
370
] | Detected guy at location [451, 103, 562, 370] | Detected Guy at location [448, 102, 566, 373]
Detected Guy at location [326, 107, 439, 353]
Detected Guy at location [219, 106, 327, 358]
Detected Guy at location [112, 101, 213, 367]
Detected guy at location [451, 103, 562, 370] | [
451,
103,
562,
370
] |
COCO_train2014_000000566368.jpg | refcoco_images/COCO_train2014_000000566368.jpg | Cellphones top row second from right | [,93.06,,,1.03,260.18,247.24] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Cellphone'] | Detected Cellphone at location [2, 2, 479, 639]
Detected Cellphones top row second from right at location [30, 0, 259, 349]
| [
30,
0,
259,
349
] | Detected Cellphones at location [30, 0, 259, 349] | Detected Cellphone at location [2, 2, 479, 639]
Detected Cellphones at location [30, 0, 259, 349] | [
30,
0,
259,
349
] |
COCO_train2014_000000548647.jpg | refcoco_images/COCO_train2014_000000548647.jpg | guy in black suit | [305.65,,72.11,472.47,473.54] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Black suit', 'Suit'] | Detected Black suit at location [327, 129, 475, 479]
Detected guy in black suit at location [304, 73, 472, 479]
| [
304,
73,
472,
479
] | Detected guy in black suit at location [304, 73, 472, 479] | Detected Black suit at location [327, 129, 475, 479]
Detected guy in black suit at location [304, 73, 472, 479] | [
304,
73,
472,
479
] |
COCO_train2014_000000517805.jpg | refcoco_images/COCO_train2014_000000517805.jpg | man | [310.65,198.26,527.46,474.61] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Man'] | Detected Man at location [308, 197, 530, 479]
Detected man at location [309, 198, 526, 479]
| [
309,
198,
526,
479
] | Detected man at location [309, 198, 526, 479] | Detected Man at location [308, 197, 530, 479]
Detected man at location [309, 198, 526, 479] | [
309,
198,
526,
479
] |
COCO_train2014_000000545405.jpg | refcoco_images/COCO_train2014_000000545405.jpg | finger ? right | [362.33,149.07,500.,,,371.2,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Finger'] | Detected Finger at location [130, 226, 357, 479]
Detected finger ? right at location [432, 174, 639, 479]
| [
432,
174,
639,
479
] | Detected finger at location [432, 174, 639, 479] | Detected Finger at location [130, 226, 357, 479]
Detected finger at location [432, 174, 639, 479] | [
432,
174,
639,
479
] |
COCO_train2014_000000565360.jpg | refcoco_images/COCO_train2014_000000565360.jpg | YELLOW SHIRT IN FRONT | [326.34,,,7.99,495.14,447.98] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Yellow shirt', 'Shirt'] | Detected Yellow shirt at location [332, 75, 499, 244]
Detected YELLOW SHIRT IN FRONT at location [130, 58, 229, 297]
| [
130,
58,
229,
297
] | Detected YELLOW SHIRT at location [130, 58, 229, 297] | Detected Yellow shirt at location [332, 75, 499, 244]
Detected YELLOW SHIRT at location [130, 58, 229, 297] | [
130,
58,
229,
297
] |
COCO_train2014_000000547859.jpg | refcoco_images/COCO_train2014_000000547859.jpg | boy on left | [,19.51,,72.61,242.36,421.78] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Boy'] | Detected Boy at location [393, 70, 622, 415]
Detected Boy at location [31, 74, 240, 418]
Detected boy on left at location [21, 73, 244, 422]
| [
21,
73,
244,
422
] | Detected boy at location [21, 73, 244, 422] | Detected Boy at location [393, 70, 622, 415]
Detected Boy at location [31, 74, 240, 418]
Detected boy at location [21, 73, 244, 422] | [
21,
73,
244,
422
] |
COCO_train2014_000000561653.jpg | refcoco_images/COCO_train2014_000000561653.jpg | white bus | [,85.,,,,85.,,,496.63,367.7,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bus'] | Detected Bus at location [81, 85, 493, 367]
Detected white bus at location [85, 85, 496, 367]
| [
85,
85,
496,
367
] | Detected white bus at location [85, 85, 496, 367] | Detected Bus at location [81, 85, 493, 367]
Detected white bus at location [85, 85, 496, 367] | [
85,
85,
496,
367
] |
COCO_train2014_000000545405.jpg | refcoco_images/COCO_train2014_000000545405.jpg | gray shirt upper right corner | [387.61,,,1.4,,500.,,,204.88] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Gray shirt', 'Shirt'] | Detected Gray shirt at location [502, 2, 639, 248]
Detected Shirt at location [503, 2, 639, 247]
Detected gray shirt upper right corner at location [491, 0, 639, 259]
| [
491,
0,
639,
259
] | Detected gray shirt at location [491, 0, 639, 259] | Detected Gray shirt at location [502, 2, 639, 248]
Detected Shirt at location [503, 2, 639, 247]
Detected gray shirt at location [491, 0, 639, 259] | [
491,
0,
639,
259
] |
COCO_train2014_000000535455.jpg | refcoco_images/COCO_train2014_000000535455.jpg | VERY TOP LEFT PIZZA | [,33.44,,,1.76,301.9,,138.65] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Pizza'] | Detected Pizza at location [39, 1, 637, 469]
| [
39,
1,
637,
469
] | Detected Pizza at location [39, 1, 637, 469] | Detected Pizza at location [39, 1, 637, 469] | [
39,
1,
637,
469
] |
COCO_train2014_000000517005.jpg | refcoco_images/COCO_train2014_000000517005.jpg | catcher | [208.25,242.77,339.42,423.41] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Catcher'] | Detected Catcher at location [200, 236, 348, 430]
Detected catcher at location [205, 242, 339, 425]
| [
205,
242,
339,
425
] | Detected catcher at location [205, 242, 339, 425] | Detected Catcher at location [200, 236, 348, 430]
Detected catcher at location [205, 242, 339, 425] | [
205,
242,
339,
425
] |
COCO_train2014_000000532734.jpg | refcoco_images/COCO_train2014_000000532734.jpg | player on left facing camera | [,46.2,,,63.53,166.37,474.59] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Player'] | Detected Player at location [195, 60, 337, 535]
Detected player on left facing camera at location [46, 64, 166, 474]
| [
46,
64,
166,
474
] | Detected player at location [46, 64, 166, 474] | Detected Player at location [195, 60, 337, 535]
Detected player at location [46, 64, 166, 474] | [
46,
64,
166,
474
] |
COCO_train2014_000000549759.jpg | refcoco_images/COCO_train2014_000000549759.jpg | left side monitor | [,,0.,,,,20.62,350.35,283.88] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Monitor'] | Detected Monitor at location [2, 20, 351, 282]
Detected left side monitor at location [0, 21, 349, 284]
| [
0,
21,
349,
284
] | Detected monitor at location [0, 21, 349, 284] | Detected Monitor at location [2, 20, 351, 282]
Detected monitor at location [0, 21, 349, 284] | [
0,
21,
349,
284
] |
COCO_train2014_000000552163.jpg | refcoco_images/COCO_train2014_000000552163.jpg | third from left holding paper | [,58.1,,207.85,123.83,477.76] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Paper'] | Detected Paper at location [23, 323, 64, 363]
Detected Paper at location [90, 342, 120, 379]
Detected third from left holding paper at location [57, 207, 123, 474]
| [
57,
207,
123,
474
] | Detected third at location [57, 207, 123, 474] | Detected Paper at location [23, 323, 64, 363]
Detected Paper at location [90, 342, 120, 379]
Detected third at location [57, 207, 123, 474] | [
57,
207,
123,
474
] |
COCO_train2014_000000533179.jpg | refcoco_images/COCO_train2014_000000533179.jpg | white bag | [186.78,133.81,350.77,439.2,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bag'] | Detected Bag at location [38, 2, 412, 262]
Detected white bag at location [186, 135, 350, 438]
| [
186,
135,
350,
438
] | Detected white bag at location [186, 135, 350, 438] | Detected Bag at location [38, 2, 412, 262]
Detected white bag at location [186, 135, 350, 438] | [
186,
135,
350,
438
] |
COCO_train2014_000000529240.jpg | refcoco_images/COCO_train2014_000000529240.jpg | man behind podium | [233.03,155.74,413.54,365.85] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Man', 'Podium'] | Detected Man at location [236, 151, 411, 362]
Detected Podium at location [73, 305, 325, 424]
Detected man behind podium at location [232, 157, 414, 366]
| [
232,
157,
414,
366
] | Detected man at location [232, 157, 414, 366] | Detected Man at location [236, 151, 411, 362]
Detected Podium at location [73, 305, 325, 424]
Detected man at location [232, 157, 414, 366] | [
232,
157,
414,
366
] |
COCO_train2014_000000561382.jpg | refcoco_images/COCO_train2014_000000561382.jpg | guy between red and white umbrellas | [368.21,,97.86,453.84,320.97] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Guy', 'Umbrella'] | Detected Guy at location [231, 97, 363, 408]
Detected Umbrella at location [238, 98, 363, 264]
Detected guy between red and white umbrellas at location [368, 98, 452, 321]
| [
368,
98,
452,
321
] | Detected guy at location [368, 98, 452, 321] | Detected Guy at location [231, 97, 363, 408]
Detected Umbrella at location [238, 98, 363, 264]
Detected guy at location [368, 98, 452, 321] | [
368,
98,
452,
321
] |
COCO_train2014_000000528130.jpg | refcoco_images/COCO_train2014_000000528130.jpg | black n white sheep | [253.8,,155.03,410.77,359.9,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Sheep'] | Detected Sheep at location [101, 149, 411, 358]
Detected Sheep at location [103, 152, 288, 333]
Detected Sheep at location [251, 157, 409, 358]
Detected black n white sheep at location [103, 152, 291, 336]
| [
103,
152,
291,
336
] | Detected black n white sheep at location [103, 152, 291, 336] | Detected Sheep at location [101, 149, 411, 358]
Detected Sheep at location [103, 152, 288, 333]
Detected Sheep at location [251, 157, 409, 358]
Detected black n white sheep at location [103, 152, 291, 336] | [
103,
152,
291,
336
] |
COCO_train2014_000000569851.jpg | refcoco_images/COCO_train2014_000000569851.jpg | right of kiwi | [272.28,285.56,421.56,427.,,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Kiwi'] | Detected Kiwi at location [56, 311, 309, 426]
Detected right of kiwi at location [272, 286, 421, 426]
| [
272,
286,
421,
426
] | Detected kiwi at location [272, 286, 421, 426] | Detected Kiwi at location [56, 311, 309, 426]
Detected kiwi at location [272, 286, 421, 426] | [
272,
286,
421,
426
] |
COCO_train2014_000000533220.jpg | refcoco_images/COCO_train2014_000000533220.jpg | left guy | [,92.53,122.17,273.74,367.96] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Guy'] | Detected Guy at location [448, 132, 637, 422]
Detected Guy at location [88, 130, 266, 391]
Detected left guy at location [88, 133, 265, 393]
| [
88,
133,
265,
393
] | Detected guy at location [88, 133, 265, 393] | Detected Guy at location [448, 132, 637, 422]
Detected Guy at location [88, 130, 266, 391]
Detected guy at location [88, 133, 265, 393] | [
88,
133,
265,
393
] |
COCO_train2014_000000526666.jpg | refcoco_images/COCO_train2014_000000526666.jpg | PERSON WITH BOOK | [232.74,,,1.5,,577.13,565.18] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Person', 'Book'] | Detected Person at location [237, 21, 579, 565]
Detected Person at location [18, 18, 199, 580]
Detected Book at location [318, 144, 473, 280]
| [
318,
144,
473,
280
] | Detected Book at location [318, 144, 473, 280] | Detected Person at location [237, 21, 579, 565]
Detected Person at location [18, 18, 199, 580]
Detected Book at location [318, 144, 473, 280] | [
318,
144,
473,
280
] |
COCO_train2014_000000553501.jpg | refcoco_images/COCO_train2014_000000553501.jpg | center bread loaf. you had that one right! | [246.5,,210.13,409.08,369.89] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bread loaf'] | Detected Bread loaf at location [3, 140, 263, 364]
Detected center bread loaf. you had that one right! at location [246, 211, 410, 372]
| [
246,
211,
410,
372
] | Detected bread loaf at location [246, 211, 410, 372] | Detected Bread loaf at location [3, 140, 263, 364]
Detected bread loaf at location [246, 211, 410, 372] | [
246,
211,
410,
372
] |
COCO_train2014_000000518361.jpg | refcoco_images/COCO_train2014_000000518361.jpg | man | [372.66,,15.21,574.21,389.78] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Man'] | Detected Man at location [375, 16, 571, 388]
Detected man at location [371, 15, 572, 389]
| [
371,
15,
572,
389
] | Detected man at location [371, 15, 572, 389] | Detected Man at location [375, 16, 571, 388]
Detected man at location [371, 15, 572, 389] | [
371,
15,
572,
389
] |
COCO_train2014_000000579785.jpg | refcoco_images/COCO_train2014_000000579785.jpg | girl in gray | [304.86,,80.3,,599.15,427.,,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Girl', 'Gray'] | Detected Girl at location [302, 81, 596, 425]
Detected girl in gray at location [304, 81, 597, 426]
| [
304,
81,
597,
426
] | Detected girl in gray at location [304, 81, 597, 426] | Detected Girl at location [302, 81, 596, 425]
Detected girl in gray at location [304, 81, 597, 426] | [
304,
81,
597,
426
] |
COCO_train2014_000000565311.jpg | refcoco_images/COCO_train2014_000000565311.jpg | red topped cookie in the bottom middle | [169.25,199.7,,354.79,363.45] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Cookie'] | Detected Cookie at location [292, 143, 638, 392]
Detected red topped cookie in the bottom middle at location [169, 200, 352, 362]
| [
169,
200,
352,
362
] | Detected red topped cookie at location [169, 200, 352, 362] | Detected Cookie at location [292, 143, 638, 392]
Detected red topped cookie at location [169, 200, 352, 362] | [
169,
200,
352,
362
] |
COCO_train2014_000000521199.jpg | refcoco_images/COCO_train2014_000000521199.jpg | the face | [,26.15,,26.15,546.,,,595.96] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Face'] | Detected Face at location [25, 27, 544, 595]
Detected Face at location [29, 76, 510, 593]
Detected the face at location [24, 27, 544, 593]
| [
24,
27,
544,
593
] | Detected face at location [24, 27, 544, 593] | Detected Face at location [25, 27, 544, 595]
Detected Face at location [29, 76, 510, 593]
Detected face at location [24, 27, 544, 593] | [
24,
27,
544,
593
] |
COCO_train2014_000000567082.jpg | refcoco_images/COCO_train2014_000000567082.jpg | bottom in front of officer | [158.39,290.48,304.45,429.,,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bottom', 'Officer'] | Detected Officer at location [202, 211, 404, 428]
Detected bottom in front of officer at location [162, 291, 304, 428]
| [
162,
291,
304,
428
] | Detected officer at location [162, 291, 304, 428] | Detected Officer at location [202, 211, 404, 428]
Detected officer at location [162, 291, 304, 428] | [
162,
291,
304,
428
] |
COCO_train2014_000000537774.jpg | refcoco_images/COCO_train2014_000000537774.jpg | chef at the bottom | [174.38,201.18,339.24,418.44] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Chef'] | Detected Chef at location [257, 121, 425, 424]
Detected chef at the bottom at location [162, 201, 344, 424]
| [
162,
201,
344,
424
] | Detected chef at location [162, 201, 344, 424] | Detected Chef at location [257, 121, 425, 424]
Detected chef at location [162, 201, 344, 424] | [
162,
201,
344,
424
] |
COCO_train2014_000000558166.jpg | refcoco_images/COCO_train2014_000000558166.jpg | person taking picture | [387.24,185.53,550.11,446.56] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Person', 'Picture'] | Detected Person at location [479, 206, 639, 479]
Detected Person at location [380, 178, 553, 461]
Detected Picture at location [2, 3, 639, 479]
Detected person taking picture at location [382, 182, 552, 479]
| [
382,
182,
552,
479
] | Detected person at location [382, 182, 552, 479] | Detected Person at location [479, 206, 639, 479]
Detected Person at location [380, 178, 553, 461]
Detected Picture at location [2, 3, 639, 479]
Detected person at location [382, 182, 552, 479] | [
382,
182,
552,
479
] |
COCO_train2014_000000533004.jpg | refcoco_images/COCO_train2014_000000533004.jpg | clock facing us on right | [298.04,,75.56,464.26,299.09] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Clock'] | Detected Clock at location [103, 62, 473, 384]
Detected Clock at location [112, 72, 283, 298]
Detected Clock at location [300, 72, 464, 297]
Detected clock facing us on right at location [300, 75, 462, 298]
| [
300,
75,
462,
298
] | Detected clock at location [300, 75, 462, 298] | Detected Clock at location [103, 62, 473, 384]
Detected Clock at location [112, 72, 283, 298]
Detected Clock at location [300, 72, 464, 297]
Detected clock at location [300, 75, 462, 298] | [
300,
75,
462,
298
] |
COCO_train2014_000000523742.jpg | refcoco_images/COCO_train2014_000000523742.jpg | rt vase | [301.19,203.54,445.59,532.23] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Vase'] | Detected Vase at location [298, 196, 443, 528]
Detected rt vase at location [299, 196, 444, 529]
| [
299,
196,
444,
529
] | Detected vase at location [299, 196, 444, 529] | Detected Vase at location [298, 196, 443, 528]
Detected vase at location [299, 196, 444, 529] | [
299,
196,
444,
529
] |
COCO_train2014_000000534323.jpg | refcoco_images/COCO_train2014_000000534323.jpg | left upper cover cloth | [,,1.08,,,3.31,258.88,237.38] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Cover', 'Cloth'] | Detected Cover at location [3, 199, 600, 466]
Detected left upper cover cloth at location [0, 3, 258, 237]
| [
0,
3,
258,
237
] | Detected upper cover cloth at location [0, 3, 258, 237] | Detected Cover at location [3, 199, 600, 466]
Detected upper cover cloth at location [0, 3, 258, 237] | [
0,
3,
258,
237
] |
COCO_train2014_000000572145.jpg | refcoco_images/COCO_train2014_000000572145.jpg | GIRL PURSE ON SHOULDER | [288.15,,89.83,352.29,352.13] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Girl', 'Purse', 'Shoulder'] | Detected Girl at location [292, 90, 354, 352]
Detected Purse at location [307, 123, 382, 202]
Detected GIRL PURSE ON SHOULDER at location [402, 76, 471, 351]
| [
402,
76,
471,
351
] | Detected GIRL PURSE at location [402, 76, 471, 351] | Detected Girl at location [292, 90, 354, 352]
Detected Purse at location [307, 123, 382, 202]
Detected GIRL PURSE at location [402, 76, 471, 351] | [
402,
76,
471,
351
] |
COCO_train2014_000000575502.jpg | refcoco_images/COCO_train2014_000000575502.jpg | cake on right | [234.43,179.78,480.,,,473.17] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Cake'] | Detected Cake at location [234, 174, 479, 478]
Detected cake on right at location [234, 178, 479, 478]
| [
234,
178,
479,
478
] | Detected cake at location [234, 178, 479, 478] | Detected Cake at location [234, 174, 479, 478]
Detected cake at location [234, 178, 479, 478] | [
234,
178,
479,
478
] |
COCO_train2014_000000531910.jpg | refcoco_images/COCO_train2014_000000531910.jpg | top part of sandwich | [100.4,,,39.88,515.73,268.18] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Sandwich'] | Detected Sandwich at location [105, 36, 539, 581]
Detected Sandwich at location [109, 37, 536, 426]
Detected top part of sandwich at location [109, 40, 515, 273]
| [
109,
40,
515,
273
] | Detected sandwich at location [109, 40, 515, 273] | Detected Sandwich at location [105, 36, 539, 581]
Detected Sandwich at location [109, 37, 536, 426]
Detected sandwich at location [109, 40, 515, 273] | [
109,
40,
515,
273
] |
COCO_train2014_000000558824.jpg | refcoco_images/COCO_train2014_000000558824.jpg | woman | [110.35,167.76,311.85,423.,,] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Woman'] | Detected Woman at location [107, 167, 313, 425]
Detected woman at location [108, 167, 313, 426]
| [
108,
167,
313,
426
] | Detected woman at location [108, 167, 313, 426] | Detected Woman at location [107, 167, 313, 425]
Detected woman at location [108, 167, 313, 426] | [
108,
167,
313,
426
] |
COCO_train2014_000000543573.jpg | refcoco_images/COCO_train2014_000000543573.jpg | bunch in front left | [,,0.,,,105.34,179.49,320.22] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bunch'] | Detected Bunch at location [3, 5, 510, 479]
Detected bunch in front left at location [1, 140, 231, 406]
| [
1,
140,
231,
406
] | Detected bunch at location [1, 140, 231, 406] | Detected Bunch at location [3, 5, 510, 479]
Detected bunch at location [1, 140, 231, 406] | [
1,
140,
231,
406
] |
COCO_train2014_000000538643.jpg | refcoco_images/COCO_train2014_000000538643.jpg | person on right | [435.78,110.02,565.21,398.02] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Person'] | Detected Person at location [94, 73, 234, 409]
Detected Person at location [254, 110, 394, 400]
Detected Person at location [437, 109, 562, 401]
Detected person on right at location [436, 111, 565, 398]
| [
436,
111,
565,
398
] | Detected person at location [436, 111, 565, 398] | Detected Person at location [94, 73, 234, 409]
Detected Person at location [254, 110, 394, 400]
Detected Person at location [437, 109, 562, 401]
Detected person at location [436, 111, 565, 398] | [
436,
111,
565,
398
] |
COCO_train2014_000000562147.jpg | refcoco_images/COCO_train2014_000000562147.jpg | brown bag on right | [491.62,198.49,639.34,356.16] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Brown bag'] | Detected Brown bag at location [195, 252, 430, 435]
Detected brown bag on right at location [492, 198, 639, 355]
| [
492,
198,
639,
355
] | Detected brown bag at location [492, 198, 639, 355] | Detected Brown bag at location [195, 252, 430, 435]
Detected brown bag at location [492, 198, 639, 355] | [
492,
198,
639,
355
] |
COCO_train2014_000000579179.jpg | refcoco_images/COCO_train2014_000000579179.jpg | blue player on the right with arm up | [311.1,,,16.07,477.56,489.04] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Blue player', 'Arm'] | Detected Blue player at location [44, 95, 372, 497]
Detected arm at location [44, 147, 194, 190]
Detected blue player on the right with arm up at location [311, 19, 474, 489]
| [
311,
19,
474,
489
] | Detected blue player at location [311, 19, 474, 489] | Detected Blue player at location [44, 95, 372, 497]
Detected arm at location [44, 147, 194, 190]
Detected blue player at location [311, 19, 474, 489] | [
311,
19,
474,
489
] |
COCO_train2014_000000528851.jpg | refcoco_images/COCO_train2014_000000528851.jpg | turned back to us kid | [446.09,,48.73,571.71,349.55] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Kid'] | Detected Kid at location [121, 33, 281, 292]
Detected Kid at location [342, 61, 449, 321]
Detected turned back to us kid at location [446, 52, 569, 347]
| [
446,
52,
569,
347
] | Detected kid at location [446, 52, 569, 347] | Detected Kid at location [121, 33, 281, 292]
Detected Kid at location [342, 61, 449, 321]
Detected kid at location [446, 52, 569, 347] | [
446,
52,
569,
347
] |
COCO_train2014_000000577657.jpg | refcoco_images/COCO_train2014_000000577657.jpg | glass left | [,,1.51,,35.89,,99.39,480.13] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Glass'] | Detected Glass at location [3, 16, 535, 638]
Detected glass left at location [0, 42, 166, 638]
| [
0,
42,
166,
638
] | Detected glass at location [0, 42, 166, 638] | Detected Glass at location [3, 16, 535, 638]
Detected glass at location [0, 42, 166, 638] | [
0,
42,
166,
638
] |
COCO_train2014_000000562675.jpg | refcoco_images/COCO_train2014_000000562675.jpg | green heart | [289.78,,87.32,492.25,297.46] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Heart'] | Detected Heart at location [290, 90, 494, 297]
Detected green heart at location [290, 88, 491, 297]
| [
290,
88,
491,
297
] | Detected green heart at location [290, 88, 491, 297] | Detected Heart at location [290, 90, 494, 297]
Detected green heart at location [290, 88, 491, 297] | [
290,
88,
491,
297
] |
COCO_train2014_000000535852.jpg | refcoco_images/COCO_train2014_000000535852.jpg | peron on right | [527.38,,50.93,637.22,426.79] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Peron'] | Detected Peron at location [0, 0, 639, 426]
Detected peron on right at location [528, 51, 636, 426]
| [
528,
51,
636,
426
] | Detected peron at location [528, 51, 636, 426] | Detected Peron at location [0, 0, 639, 426]
Detected peron at location [528, 51, 636, 426] | [
528,
51,
636,
426
] |
COCO_train2014_000000522256.jpg | refcoco_images/COCO_train2014_000000522256.jpg | kid holding bat | [,84.11,,59.41,295.45,421.72] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Kid', 'Bat'] | Detected Kid at location [84, 61, 299, 425]
Detected Kid at location [373, 122, 506, 480]
Detected kid holding bat at location [84, 63, 297, 423]
| [
84,
63,
297,
423
] | Detected kid at location [84, 63, 297, 423] | Detected Kid at location [84, 61, 299, 425]
Detected Kid at location [373, 122, 506, 480]
Detected kid at location [84, 63, 297, 423] | [
84,
63,
297,
423
] |
COCO_train2014_000000550476.jpg | refcoco_images/COCO_train2014_000000550476.jpg | pony tailed dad | [247.69,,35.22,500.,,,328.56] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Pony tailed dad', 'Dad'] | Detected Pony tailed dad at location [307, 43, 638, 423]
Detected Dad at location [307, 43, 637, 423]
Detected pony tailed dad at location [312, 46, 638, 425]
| [
312,
46,
638,
425
] | Detected pony tailed dad at location [312, 46, 638, 425] | Detected Pony tailed dad at location [307, 43, 638, 423]
Detected Dad at location [307, 43, 637, 423]
Detected pony tailed dad at location [312, 46, 638, 425] | [
312,
46,
638,
425
] |
COCO_train2014_000000579156.jpg | refcoco_images/COCO_train2014_000000579156.jpg | cow on the far right | [402.42,140.32,500.,,,342.74] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Cow'] | Detected Cow at location [56, 170, 639, 431]
Detected Cow at location [59, 186, 384, 403]
Detected cow on the far right at location [514, 179, 639, 435]
| [
514,
179,
639,
435
] | Detected cow at location [514, 179, 639, 435] | Detected Cow at location [56, 170, 639, 431]
Detected Cow at location [59, 186, 384, 403]
Detected cow at location [514, 179, 639, 435] | [
514,
179,
639,
435
] |
COCO_train2014_000000532577.jpg | refcoco_images/COCO_train2014_000000532577.jpg | child | [222.47,171.01,389.32,445.05] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Child'] | Detected Child at location [227, 173, 390, 450]
Detected child at location [226, 172, 389, 450]
| [
226,
172,
389,
450
] | Detected child at location [226, 172, 389, 450] | Detected Child at location [227, 173, 390, 450]
Detected child at location [226, 172, 389, 450] | [
226,
172,
389,
450
] |
COCO_train2014_000000519911.jpg | refcoco_images/COCO_train2014_000000519911.jpg | middle bear | [,61.52,,33.67,365.77,478.86] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bear'] | Detected Bear at location [82, 36, 462, 616]
Detected middle bear at location [83, 36, 462, 617]
| [
83,
36,
462,
617
] | Detected bear at location [83, 36, 462, 617] | Detected Bear at location [82, 36, 462, 616]
Detected bear at location [83, 36, 462, 617] | [
83,
36,
462,
617
] |
COCO_train2014_000000544113.jpg | refcoco_images/COCO_train2014_000000544113.jpg | right small bowl | [271.79,410.52,397.13,559.96] | <|im_start|>system
You are a coordinate selection system. Your task is to:
1. Parse the given bounding box coordinates
2. For left/right selection:
- Compare the starting x-coordinates (first number in each box)
- Left means smaller x-coordinate
- Right means larger x-coordinate
3. For top/bottom selection:
... | ['Bowl'] | Detected Bowl at location [270, 409, 400, 562]
Detected right small bowl at location [270, 409, 400, 563]
| [
270,
409,
400,
563
] | Detected small bowl at location [270, 409, 400, 563] | Detected Bowl at location [270, 409, 400, 562]
Detected small bowl at location [270, 409, 400, 563] | [
270,
409,
400,
563
] |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 8