| # | key | jpg | json |
|---|
1
|
000000000
|
|
{
"caption": "The image displays a <object>white sliding door frame</object><bbox>[282, 19, 718, 994]</bbox> constructed from a material with a grainy texture on the sides and a smooth finish on top. It features two <object>frosted glass sliding doors</object><bbox>[316, 57, 491, 953], <bbox>[523, 57, 698, 953]</bbox>, each with a <object>window-like opening</object><bbox>[330, 67, 477, 942], <bbox>[537, 67, 684, 942]</bbox> in the center that allows light to pass through while maintaining privacy. The doors are positioned side by side within the frame, creating a symmetrical appearance suitable for closet or room dividers.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000000"
}
|
2
|
000000001
|
|
{
"caption": "The image displays a set of automotive accessories arranged on a <object>white background</object><bbox>[0, 0, 1000, 1000]</bbox>, featuring two large, curved <object>black fender flares</object><bbox>[10, 10, 688, 458]</bbox> positioned at the top and another pair of <object>black fender flares</object><bbox>[300, 481, 992, 991]</bbox> at the bottom right. To the left, there is a <object>coil of black rubber tubing</object><bbox>[20, 527, 181, 703]</bbox> and a small collection of <object>metal bolts and nuts</object><bbox>[196, 670, 297, 733]</bbox>. Below these items, there are organized rows of <object>silver washers</object><bbox>[25, 719, 302, 850]</bbox> and <object>silver screws</object><bbox>[25, 867, 255, 985]</bbox>, all neatly laid out to showcase the complete kit for vehicle modification.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000001"
}
|
3
|
000000002
|
|
{
"caption": "The image features a <object>reclaimed wood dining table</object><bbox>[38, 175, 964, 842]</bbox> with a <object>rectangular tabletop</object><bbox>[38, 175, 964, 365]</bbox> displaying a natural wood grain and rustic finish. The table is supported by two <object>whitewashed pedestal bases</object><bbox>[347, 342, 678, 842]</bbox><bbox>[597, 337, 826, 587]</bbox> that resemble stone or concrete, adding to its rustic charm. The table is placed on a <object>hardwood floor</object><bbox>[0, 295, 1000, 1000]</bbox> with a rich, warm tone. In the background, there is a <object>white wall</object><bbox>[272, 0, 1000, 425]</bbox> and a <object>doorway</object><bbox>[147, 0, 214, 180]</bbox> leading to another room. A <object>black chair</object><bbox>[0, 135, 100, 515]</bbox> is partially visible on the left side of the image, and a <object>white radiator</object><bbox>[304, 105, 397, 205]</bbox> is mounted on the wall in the background.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000002"
}
|
4
|
000000003
|
|
{
"caption": "The image displays a modern bathroom featuring a <object>glass shower enclosure</object><bbox>[0, 72, 485, 972]</bbox> with a <object>chrome showerhead</object><bbox>[318, 262, 368, 296]</bbox> and <object>shower controls</object><bbox>[98, 468, 165, 533]</bbox> mounted on a <object>beige tiled wall</object><bbox>[0, 0, 440, 735]</bbox>. Inside the shower, a <object>dark brown shower tray</object><bbox>[25, 704, 475, 972]</bbox> contains a <object>circular drain</object><bbox>[265, 822, 320, 850]</bbox>. Adjacent to the shower is a <object>dark brown cabinet</object><bbox>[485, 543, 605, 848]</bbox> topped with two <object>tall red vases</object><bbox>[528, 380, 585, 543]</bbox>, situated beneath a <object>dark brown overhead cabinet</object><bbox>[485, 16, 605, 232]</bbox> with a <object>recessed light</object><bbox>[555, 192, 590, 213]</bbox>.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000003"
}
|
5
|
000000004
|
|
{
"caption": "The image features a close-up view of a person's face, focusing on the lower half, including the <object>mouth</object><bbox>[276, 289, 432, 408]</bbox> and <object>chin</object><bbox>[276, 408, 432, 600]</bbox>. The individual is holding their <object>hand</object><bbox>[381, 200, 741, 1000]</bbox> against their <object>cheek</object><bbox>[432, 0, 800, 450]</bbox>, with fingers spread out and the thumb resting near the <object>jawline</object><bbox>[432, 200, 741, 450]</bbox>. A noticeable <object>redness</object><bbox>[500, 100, 800, 450]</bbox> is visible on the cheek, suggesting discomfort or pain in that area. The background is a plain, light color, ensuring the focus remains on the person's facial expression and the gesture indicating distress.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000004"
}
|
6
|
000000006
|
|
{
"caption": "The image displays a hotel room featuring two <object>beds</object><bbox>[0, 676, 472, 1000]</bbox><bbox>[102, 456, 817, 990]</bbox> covered with colorful patchwork quilts and topped with <object>white pillows</object><bbox>[145, 464, 375, 640]</bbox>. The room has light blue walls, with a <object>wooden headboard</object><bbox>[97, 456, 375, 632]</bbox> behind the main bed and a <object>wooden chair</object><bbox>[375, 484, 472, 616]</bbox> tucked beside it. Lighting is provided by a <object>wall-mounted lamp</object><bbox>[165, 144, 250, 460]</bbox> on the left wall and a <object>floor lamp</object><bbox>[475, 300, 542, 604]</bbox> in the corner. A <object>framed picture</object><bbox>[380, 156, 425, 292]</bbox> hangs above the chair, while a <object>brown curtain</object><bbox>[590, 100, 695, 544]</bbox> partially covers one of the two <object>large windows</object><bbox>[695, 104, 970, 520]</bbox> that let in natural light.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000006"
}
|
7
|
000000007
|
|
{
"caption": "The image features a vintage <object>biplane</object><bbox>[0, 0, 1000, 909]</bbox> prominently displaying the text \"<object>Varney air lines</object><bbox>[520, 387, 732, 460]</bbox>\" and \"<object>U.S. AIR MAIL</object><bbox>[524, 484, 710, 553]</bbox>\" on its fuselage. A <object>pilot</object><bbox>[491, 150, 605, 284]</bbox> wearing a cap and goggles sits in the open cockpit, while three men stand on the <object>ground</object><bbox>[0, 574, 1000, 1000]</bbox> below. On the left, a <object>man in a suit and hat</object><bbox>[147, 216, 277, 927]</bbox> observes the scene, and in the center, another <object>man in a suit</object><bbox>[370, 294, 542, 868]</bbox> gestures toward the aircraft. To the right, a third <object>man in a suit</object><bbox>[714, 275, 843, 934]</bbox> stands holding a hat, facing the other two men. The plane's <object>propeller</object><bbox>[26, 193, 138, 400]</bbox> is visible at the front, along with its <object>wheels</object><bbox>[0, 583, 56, 787]</bbox>",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000007"
}
|
8
|
000000008
|
|
{
"caption": "The image features a <object>martial artist</object><bbox>[0, 88, 918, 1000]</bbox> in a dynamic stance, wearing a <object>white gi</object><bbox>[0, 326, 876, 1000]</bbox> with a <object>black belt</object><bbox>[262, 712, 543, 941]</bbox> tied around the waist. The artist has <object>short brown hair</object><bbox>[415, 88, 566, 297]</bbox> and is depicted with an intense facial expression, showing <object>furrowed brows</object><bbox>[476, 193, 552, 236]</bbox> and an <object>open mouth</object><bbox>[505, 288, 534, 326]</bbox>. One <object>hand</object><bbox>[135, 577, 248, 723]</bbox> is clenched into a fist near the chest, while the other <object>hand</object><bbox>[763, 319, 918, 525]</bbox> is extended outward with fingers spread, suggesting a defensive or striking position. The background is a plain <object>light gray gradient</object><bbox>[0, 0, 1000, 1000]</bbox>, emphasizing the figure.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000008"
}
|
9
|
000000009
|
|
{
"caption": "The image displays a pair of <object>dangle earrings</object><bbox>[0, 0, 850, 1000]</bbox> resting on a <object>white surface</object><bbox>[0, 0, 1000, 1000]</bbox>. The left earring features a <object>gold lever-back clasp</object><bbox>[0, 0, 360, 300]</bbox> connected to a delicate <object>gold chain</object><bbox>[245, 280, 400, 650]</bbox> that leads down to a <object>gold cap with clear crystals</object><bbox>[338, 600, 490, 820]</bbox> and a large, round <object>blue lapis lazuli bead</object><bbox>[345, 750, 610, 1000]</bbox>. The right earring shows a similar structure with a <object>gold cap with clear crystals</object><bbox>[590, 210, 740, 390]</bbox> sitting atop a matching <object>blue lapis lazuli bead</object><bbox>[600, 330, 850, 650]</bbox>, which has a small <object>gold post</object><bbox>[750, 590, 795, 650]</bbox> at its base.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000009"
}
|
10
|
000000011
|
|
{
"caption": "The image features a <object>black background</object><bbox>[0, 0, 1000, 1000]</bbox> that highlights a large <object>circular stained glass window</object><bbox>[150, 100, 850, 550]</bbox> at the top, which is composed of numerous <object>smaller circular panels</object><bbox>[180, 130, 820, 520]</bbox> filled with intricate designs and vibrant colors. Below the circular window, there are five <object>vertical stained glass panels</object><bbox>[150, 580, 850, 880]</bbox>, each containing detailed imagery and rich hues, arranged in a row. The overall composition suggests a religious or historical theme, typical of cathedral architecture.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000011"
}
|
11
|
000000012
|
|
{
"caption": "The image displays a digital interface featuring a <object>gray rectangular area</object><bbox>[100, 150, 850, 750]</bbox> that serves as the main content section, surrounded by a <object>white border</object><bbox>[50, 100, 900, 850]</bbox>. At the top of the interface, there is a <object>navigation bar</object><bbox>[100, 50, 850, 150]</bbox> containing various <object>icons</object><bbox>[150, 70, 800, 130]</bbox> for different functionalities. Below the main content area, there are additional <object>text elements</object><bbox>[100, 750, 850, 850]</bbox> providing further information or instructions. The overall layout is clean and organized, with distinct sections for navigation, content, and supplementary details.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000012"
}
|
12
|
000000014
|
|
{
"caption": "The image displays an interior view of a room featuring a <object>white door</object><bbox>[382, 122, 824, 870]</bbox> with a large <object>frosted glass panel</object><bbox>[416, 172, 772, 812]</bbox> that obscures the view outside. To the left of the door, there is a <object>window</object><bbox>[0, 127, 350, 650]</bbox> offering a glimpse of the outdoors, where a <object>swing set</object><bbox>[66, 362, 286, 597]</bbox> and some <object>greenery</object><bbox>[0, 302, 344, 592]</bbox> are visible. In the foreground, a portion of a <object>couch</object><bbox>[0, 652, 404, 1000]</bbox> with a <object>blue and white patterned cushion</object><bbox>[0, 652, 304, 902]</bbox> can be seen. The floor is covered with a <object>dark rug</object><bbox>[474, 882, 824, 1000]</bbox> in front of the door, and the walls are painted in a <object>light color</object><bbox>[0, 0, 1000, 1000]</bbox>, contributing to a bright and airy atmosphere.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000014"
}
|
13
|
000000015
|
|
{
"caption": "The image features a <object>white electronic lighter</object><bbox>[347, 283, 500, 991]</bbox> positioned vertically against a <object>dark background</object><bbox>[0, 0, 1000, 1000]</bbox>. The lighter has a <object>rectangular body</object><bbox>[347, 283, 500, 991]</bbox> with a <object>small window</object><bbox>[403, 500, 445, 700]</bbox> on the front, likely for displaying battery status or activation. At the top, there is a <object>metallic ignition area</object><bbox>[380, 283, 470, 350]</bbox> emitting a <object>blue plasma arc</object><bbox>[390, 0, 460, 280]</bbox>, indicating its windproof and flameless ignition capability. Diagonally crossing behind the lighter are two <object>white rods</object><bbox>[0, 0, 1000, 650]</bbox>, one of which has an <object>orange tip</object><bbox>[850, 500, 1000, 650]</bbox>, possibly representing a cigarette or another object being lit.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000015"
}
|
14
|
000000016
|
|
{
"caption": "The image displays a <object>black cylindrical tube</object><bbox>[127, 300, 878, 674]</bbox> lying horizontally on a <object>white surface</object><bbox>[0, 0, 1000, 1000]</bbox>. The tube features <object>white text</object><bbox>[183, 385, 698, 531]</bbox> printed along its side, which reads \"4-1/2\" FIT PUC4138 #FITBIKECY\". Additionally, there is a <object>small white label</object><bbox>[127, 365, 167, 422]</bbox> with a barcode or similar marking near the left end of the tube. The <object>open end</object><bbox>[700, 385, 878, 674]</bbox> of the tube is visible on the right side, showing the hollow interior.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000016"
}
|
15
|
000000017
|
|
{
"caption": "The image features a <object>red apple</object><bbox>[50, 333, 350, 617]</bbox> resting on top of a <object>stack of books</object><bbox>[0, 600, 750, 950]</bbox>, with a <object>human hand</object><bbox>[500, 417, 900, 783]</bbox> reaching toward the books from the right side. The scene is set against a <object>white background</object><bbox>[0, 0, 1000, 1000]</bbox>, which highlights the vibrant colors of the apple and the varied hues of the book spines.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000017"
}
|
16
|
000000018
|
|
{
"caption": "The image showcases a vibrant <object>outdoor summer BBQ setup</object><bbox>[0, 532, 1000, 966]</bbox> featuring a gourmet burger bar, with a <object>wooden table</object><bbox>[0, 654, 1000, 966]</bbox> laden with various food items. In the foreground, there are <object>white rectangular plates</object><bbox>[10, 734, 312, 868]</bbox> holding <object>yellow cheese slices</object><bbox>[100, 750, 250, 850]</bbox> and <object>raw meat patties</object><bbox>[340, 734, 600, 850]</bbox>. To the right, another <object>white plate</object><bbox>[630, 740, 870, 870]</bbox> displays <object>green vegetables</object><bbox>[650, 750, 800, 850]</bbox> and <object>red tomato slices</object><bbox>[680, 760, 780, 830]</bbox>. Behind these, a <object>tiered stand</object><bbox>[550, 460, 820, 700]</bbox> holds an assortment of <object>burgers</object><bbox>[580, 600, 750, 680]</bbox> and <object>condiments</object><bbox>[580, 650, 750, 700]</bbox>.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000018"
}
|
17
|
000000019
|
|
{
"caption": "The image features a <object>red rectangular box</object><bbox>[12, 150, 957, 850]</bbox> prominently displaying the brand name \"<object>Brillo</object><bbox>[57, 375, 520, 675]</bbox>\" in bold blue and red letters on its front face. The top surface of the box also shows the \"<object>Brillo</object><bbox>[140, 170, 600, 330]</bbox>\" logo in black and red text, alongside the description \"<object>RUST-FREE SOAP PADS</object><bbox>[610, 250, 820, 390]</bbox>\" in white lettering. On the right side panel, the text \"<object>10 HOTEL SIZE SOAP PADS</object><bbox>[550, 490, 770, 760]</bbox>\" is printed in white against a dark blue background, while the left side panel partially reveals the word \"<object>Brillo</object><bbox>[810, 570, 910, 770]</bbox>\" in red. The entire package rests against a plain <object>white background</object><bbox>[0, 0, 1000, 1000]</bbox>, highlighting the vibrant colors and typography of the product packaging.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000019"
}
|
18
|
000000021
|
|
{
"caption": "The image displays a <object>pretend play set</object><bbox>[0, 200, 442, 988]</bbox> designed for children, prominently featuring a <object>blue passport</object><bbox>[312, 10, 548, 320]</bbox> with a gold emblem at the top. Below the passport is a <object>sticker sheet</object><bbox>[432, 120, 918, 570]</bbox> filled with colorful travel-themed stickers, including images of landmarks and animals. To the left, there is a <object>cardboard backing</object><bbox>[0, 200, 442, 988]</bbox> with the text \"Pretend Play Building Imaginations World Traveler\" and various illustrations such as a zebra and a globe. In front of these items lies a <object>ticket</object><bbox>[440, 510, 990, 750]</bbox> resembling a boarding pass with sections for details like destination and date. Additionally, there is a <object>postcard</object><bbox>[498, 650, 942, 980]</bbox> depicting the Golden Gate Bridge with the text \"San Francisco.\" The overall arrangement suggests an educational toy aimed at fostering imaginative play related to world travel.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000021"
}
|
19
|
000000022
|
|
{
"caption": "The image captures a meticulously maintained park scene in Schenley Park during the year 1916, featuring an exquisitely designed <object>flower bed</object><bbox>[38, 600, 964, 850]</bbox> in the foreground with intricate patterns and encircled by neatly trimmed <object>hedges</object><bbox>[38, 600, 964, 850]</bbox>. A <object>pathway</object><bbox>[0, 550, 1000, 1000]</bbox> leads to a serene grove of evenly spaced tall <object>trees</object><bbox>[0, 200, 1000, 500]</bbox>, creating a harmonious blend of nature and design. The background reveals a <object>grassy field</object><bbox>[0, 200, 1000, 500]</bbox> bordered by a dense line of <object>trees</object><bbox>[0, 200, 1000, 500]</bbox>, enhancing the tranquil atmosphere. The overall composition highlights the careful landscaping and the timeless beauty of the park's layout.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000022"
}
|
20
|
000000023
|
|
{
"caption": "The image features a <object>mannequin</object><bbox>[24, 0, 976, 1000]</bbox> standing against a <object>white background</object><bbox>[0, 0, 1000, 1000]</bbox>, wearing a <object>striped shirt dress</object><bbox>[31, 65, 969, 958]</bbox>. The dress is characterized by its <object>long sleeves</object><bbox>[31, 138, 223, 558]</bbox><bbox>[777, 142, 969, 562]</bbox> and a <object>collar</object><bbox>[250, 65, 750, 288]</bbox>, with a <object>buttoned front</object><bbox>[456, 288, 543, 958]</bbox> that runs down the center. The fabric displays a pattern of <object>vertical stripes</object><bbox>[31, 65, 969, 958]</bbox> in shades of blue and white, giving it a loose, robe-like appearance.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000023"
}
|
21
|
000000024
|
|
{
"caption": "The image showcases a luxurious walk-in shower area featuring walls entirely covered in glossy, dark blue subway tiles <bbox>[0, 0, 1000, 1000]</bbox> with white grout lines. A sleek, brass-finished showerhead <bbox>[210, 218, 447, 660]</bbox> is mounted on the left wall, accompanied by a matching handheld shower wand and control knobs <bbox>[0, 522, 175, 788]</bbox>. The shower includes a built-in bench <bbox>[672, 762, 1000, 954]</bbox> on the right side, also clad in the same blue tiles, providing a comfortable seating area. A recessed niche <bbox>[401, 438, 582, 561]</bbox> in the back wall offers storage space, while a circular soap dish <bbox>[608, 285, 682, 342]</bbox> is mounted above it. The floor features a contrasting beige stone tile <bbox>[323, 905, 817, 1000]</bbox>, and a recessed ceiling light <bbox>[590, 0, 693, 27]</bbox> illuminates the space from above.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000024"
}
|
22
|
000000025
|
|
{
"caption": "The image displays three <object>computer cooling fans</object><bbox>[104, 335, 820, 728]</bbox> arranged horizontally on a weathered <object>wooden deck surface</object><bbox>[0, 0, 1000, 1000]</bbox>. The fan on the left features a <object>yellow center label</object><bbox>[168, 458, 246, 604]</bbox>, the middle fan has a <object>white center label</object><bbox>[416, 458, 494, 604]</bbox>, and the rightmost fan displays a <object>blue center label</object><bbox>[672, 458, 750, 604]</bbox>. Each fan is equipped with <object>black plastic frames</object><bbox>[104, 335, 820, 728]</bbox> and <object>white wire grilles</object><bbox>[104, 335, 820, 728]</bbox>, with <object>colored electrical wires</object><bbox>[233, 0, 811, 850]</bbox> extending from them. Overlaid on the image are large white text elements reading \"<object>SOLAR POWERED</object><bbox>[126, 130, 826, 250]</bbox>\" at the top and \"<object>RV VENT FAN</object><bbox>[212, 825, 745, 940]</bbox>\" at the bottom.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000025"
}
|
23
|
000000026
|
|
{
"caption": "The image displays a children's outfit featuring a <object>light pink long-sleeved top</object><bbox>[5, 7, 991, 472]</bbox> with a <object>black polka-dotted bow</object><bbox>[250, 275, 744, 398]</bbox> at the waist. Below the top is a <object>black skirt with white polka dots</object><bbox>[102, 385, 846, 646]</bbox> adorned with three <object>pink heart-shaped appliqu\u00e9s</object><bbox>[127, 461, 823, 618]</bbox>. The ensemble is completed with <object>black and white striped leggings</object><bbox>[241, 632, 682, 991]</bbox> that have small <object>pink bows</object><bbox>[231, 940, 672, 992]</bbox> at the ankles.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000026"
}
|
24
|
000000027
|
|
{
"caption": "The underwater scene is dominated by a dense school of small, silvery fish <bbox>[0, 0, 1000, 780]</bbox> swimming in the blue water <bbox>[0, 0, 1000, 1000]</bbox>. Sunlight filters through the rippled surface <bbox>[0, 0, 1000, 250]</bbox>, creating shimmering patterns and illuminating the aquatic environment. The bottom of the frame features a rocky seabed <bbox>[0, 680, 1000, 1000]</bbox> composed of large, dark stones and boulders <bbox>[370, 810, 540, 1000]</bbox><bbox>[620, 800, 900, 1000]</bbox> that anchor the composition.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000027"
}
|
25
|
000000029
|
|
{
"caption": "The image showcases a beautifully crafted <object>wooden hamper</object><bbox>[288, 250, 865, 803]</bbox> with intricate designs, open to reveal its contents. Inside the hamper, there is a generous amount of <object>almonds</object><bbox>[326, 550, 570, 726]</bbox>, neatly arranged alongside a <object>decorative card</object><bbox>[517, 590, 676, 720]</bbox> and several <object>chocolates</object><bbox>[644, 513, 953, 733]</bbox> wrapped in dark foil. To the left of the hamper, a <object>traditional rakhi</object><bbox>[166, 473, 222, 736]</bbox> is displayed on a <object>card</object><bbox>[76, 463, 294, 753]</bbox>, accompanied by small decorative items such as <object>coins</object><bbox>[144, 713, 276, 803]</bbox>. The background features a softly blurred <object>kitchen setting</object><bbox>[0, 0, 1000, 476]</bbox> with various household items, including a <object>bowl</object><bbox>[0, 173, 153, 373]</bbox> and <object>decorative objects</object><bbox>[236, 0, 426, 173]</bbox>, adding to the homely ambiance.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000029"
}
|
26
|
000000030
|
|
{
"caption": "The image displays a <object>black laptop power adapter</object><bbox>[118, 250, 775, 545]</bbox> featuring a <object>circular white label</object><bbox>[312, 338, 443, 472]</bbox> on its top surface, nestled within a <object>clear plastic clamshell package</object><bbox>[22, 115, 977, 915]</bbox>. Surrounding the main unit are several <object>black power cables</object><bbox>[35, 165, 935, 865]</bbox>, including a coiled cord at the bottom and another cable extending from the top right. The packaging rests on a <object>white background</object><bbox>[0, 0, 1000, 1000]</bbox>, with a small <object>yellow text label</object><bbox>[735, 835, 955, 885]</bbox> visible in the bottom right corner of the plastic case.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000030"
}
|
27
|
000000031
|
|
{
"caption": "The image displays a <object>transparent plastic case</object><bbox>[118, 12, 909, 970]</bbox> with a <object>hinged lid</object><bbox>[312, 12, 909, 494]</bbox> that is open, revealing its contents against a <object>black background</object><bbox>[0, 0, 1000, 1000]</bbox>. Inside the case rests a <object>memory card holder</object><bbox>[250, 459, 714, 831]</bbox> featuring a distinct <object>red top section</object><bbox>[394, 467, 706, 620]</bbox>, a <object>black middle section</object><bbox>[272, 536, 693, 793]</bbox>, and a <object>white bottom strip</object><bbox>[297, 631, 514, 762]</bbox>. The case itself has a <object>locking latch</object><bbox>[172, 772, 446, 960]</bbox> on the front edge and molded indentations designed to securely hold the card adapter.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000031"
}
|
28
|
000000033
|
|
{
"caption": "The image captures a somber memorial scene on a <object>paved ground</object><bbox>[0, 270, 1000, 1000]</bbox> where a <object>man in a white shirt</object><bbox>[45, 166, 540, 926]</bbox> kneels to place <object>flowers</object><bbox>[338, 653, 466, 894]</bbox> among numerous <object>red candles</object><bbox>[528, 384, 997, 807]</bbox>. Beside him, a <object>woman in a black dress</object><bbox>[354, 0, 620, 576]</bbox> crouches to arrange items near a <object>brown teddy bear</object><bbox>[532, 403, 618, 520]</bbox>. A large <object>white sign</object><bbox>[553, 384, 960, 781]</bbox> displays handwritten messages in Spanish, while a framed <object>photograph</object><bbox>[385, 788, 543, 1000]</bbox> and a <object>Spanish flag</object><bbox>[862, 756, 1000, 904]</bbox> lie nearby.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000033"
}
|
29
|
000000034
|
|
{
"caption": "The image displays a <object>gray border</object><bbox>[0, 0, 1000, 1000]</bbox> framing a central <object>yellow background</object><bbox>[52, 71, 946, 928]</bbox> that radiates from a bright white center. Multiple arms of varying skin tones extend toward the middle: a <object>light beige arm</object><bbox>[182, 71, 526, 484]</bbox> from the top left, a <object>dark brown arm</object><bbox>[494, 71, 794, 442]</bbox> from the top right, a <object>reddish-brown arm</object><bbox>[514, 418, 946, 662]</bbox> from the right, a <object>medium tan arm</object><bbox>[514, 500, 760, 928]</bbox> from the bottom right, a <object>light peach arm</object><bbox>[408, 462, 518, 928]</bbox> from the bottom center, a <object>dark brown arm</object><bbox>[52, 508, 416, 906]</bbox> from the bottom left, and a <object>pale yellow arm</object><bbox>[52, 348, 436, 526]</bbox> from the left. Their <object>hands</object><bbox>[246, 296, 706, 682]</bbox> overlap in a unified stack at the center, symbolizing diversity and cooperation.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000034"
}
|
30
|
000000035
|
|
{
"caption": "The image displays a <object>Coach handbag</object><bbox>[78, 113, 932, 951]</bbox> featuring a <object>beige canvas body</object><bbox>[120, 382, 870, 948]</bbox> adorned with a repeating <object>signature \"C\" pattern</object><bbox>[130, 400, 860, 930]</bbox> in a darker tan hue. The bag is accented with vibrant <object>pink leather trim</object><bbox>[78, 113, 932, 771]</bbox> along the top edge and handles, including a <object>pink shoulder strap</object><bbox>[78, 390, 478, 771]</bbox> attached via <object>gold-tone hardware</object><bbox>[158, 290, 208, 400]</bbox><bbox>[808, 290, 858, 400]</bbox>. A <object>pink leather tag</object><bbox>[808, 420, 878, 560]</bbox> hangs from the side handle, while a <object>gold-tone oval logo plaque</object><bbox>[478, 820, 578, 870]</bbox> is centered on the front panel. The bag rests against a plain <object>white background</object><bbox>[0, 0, 1000, 1000]</bbox>, highlighting its structured yet slouchy hobo silhouette.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000035"
}
|
31
|
000000036
|
|
{
"caption": "The image displays a close-up view of several <object>bacon-wrapped dates</object><bbox>[214, 383, 572, 772]</bbox><bbox>[471, 526, 792, 988]</bbox><bbox>[703, 592, 981, 997]</bbox><bbox>[350, 190, 696, 545]</bbox><bbox>[757, 336, 999, 714]</bbox><bbox>[457, 0, 732, 223]</bbox><bbox>[750, 0, 999, 333]</bbox> arranged on a <object>white plate</object><bbox>[0, 0, 1000, 1000]</bbox>. Each date is encased in <object>crispy bacon</object><bbox>[214, 383, 572, 772]</bbox> and pierced with a <object>wooden toothpick</object><bbox>[407, 45, 471, 436]</bbox><bbox>[563, 0, 640, 538]</bbox><bbox>[837, 443, 870, 776]</bbox><bbox>[864, 19, 916, 407]</bbox> to hold the wrapping in place. The <object>dark, sticky date filling</object><bbox>[221, 506, 357, 688]</bbox>",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000036"
}
|
32
|
000000037
|
|
{
"caption": "The image features a person wearing a <object>Spider-Man costume</object><bbox>[240, 10, 790, 990]</bbox> that is predominantly <object>red</object><bbox>[240, 10, 790, 990]</bbox> and <object>blue</object><bbox>[240, 10, 790, 990]</bbox> with a detailed <object>web pattern</object><bbox>[240, 10, 790, 990]</bbox>. The <object>mask</object><bbox>[370, 10, 580, 160]</bbox> covers the entire head, featuring <object>large white eyes</object><bbox>[390, 50, 560, 100]</bbox> with black outlines. The <object>chest area</object><bbox>[300, 160, 680, 450]</bbox> displays a prominent <object>black spider emblem</object><bbox>[400, 210, 540, 300]</bbox>. The <object>arms</object><bbox>[240, 160, 360, 530]</bbox><bbox>[640, 160, 790, 530]</bbox> are primarily blue with red sections near the shoulders and wrists.",
"source_tar": "/capstor/store/cscs/swissai/infra01/vision-datasets/raw/stage2/hf___Salesforce___blip3-grounding-50m/01/chunk_000/00000.tar",
"key": "000000037"
}
|