Create OmniParser demo

[xenova]

Run with:

node index.js

Input image:

Example output:

{
  text: `A white background with the words "I'm Feeling Lucky" in black.`,
  bbox: [ 1618, 1051, 1903, 1128 ],
  score: 0.7265540361404419
}
{
  text: 'A white background with the words Google Search in black.',
  bbox: [ 1340, 1051, 1598, 1127 ],
  score: 0.687532901763916
}
{
  text: 'A stylized image of a blue circle with a green circle in the middle.',
  bbox: [ 2256, 2083, 2319, 2151 ],
  score: 0.5648190975189209
}
{
  text: 'A blue circle with the words "sign in" in the middle.',
  bbox: [ 3010, 168, 3187, 248 ],
  score: 0.5589484572410583
}
{
  text: 'A white sign that says Search and Microsoft.',
  bbox: [ 826, 2086, 1357, 2151 ],
  score: 0.5514659285545349
}
{
  text: 'A blurry image of the word business in black.',
  bbox: [ 278, 1985, 398, 2034 ],
  score: 0.5280276536941528
}
{
  text: 'A black and blue image of squares with a button in the middle.',
  bbox: [ 2172, 2085, 2233, 2151 ],
  score: 0.5184041261672974
}
{
  text: 'A blurry image of a computer screen with a black background.',
  bbox: [ 2082, 2084, 2146, 2153 ],
  score: 0.47427114844322205
}
{
  text: 'A graphic of a hand holding a tablet.',
  bbox: [ 2344, 2084, 2400, 2148 ],
  score: 0.4604111611843109
}
{
  text: 'A blue image of a person with an arrow pointing to the letter O.',
  bbox: [ 1643, 2076, 1712, 2152 ],
  score: 0.4367396831512451
}
{
  text: 'How Search Works in black and white',
  bbox: [ 452, 1985, 684, 2032 ],
  score: 0.426216185092926
}
{
  text: 'A red and blue number with the numbers T and 2.',
  bbox: [ 1992, 2082, 2065, 2152 ],
  score: 0.38590338826179504
}
{
  text: 'A white background with the words advertising in black.',
  bbox: [ 60, 1963, 236, 2053 ],
  score: 0.36276888847351074
}
{
  text: 'black white image square shaped words',
  bbox: [ 1905, 2077, 1968, 2155 ],
  score: 0.35925644636154175
}
{
  text: 'A colorful sign with a person holding a briefcase.',
  bbox: [ 1374, 2087, 1439, 2146 ],
  score: 0.3518167734146118
}
{
  text: 'A yellow square with a blue rectangle in the middle.',
  bbox: [ 1462, 2083, 1529, 2153 ],
  score: 0.34263908863067627
}
{
  text: 'A white background with the words advertising in black.',
  bbox: [ 78, 1980, 229, 2036 ],
  score: 0.3385675251483917
}
{
  text: 'A purple letter N is in the middle of a black background.',
  bbox: [ 1550, 2080, 1617, 2153 ],
  score: 0.33095934987068176
}
{
  text: 'the word settings is in black letters on a white background',
  bbox: [ 3057, 1987, 3160, 2037 ],
  score: 0.3290565609931946
}
{
  text: 'A white sign that saysprivacy in black letters.',
  bbox: [ 2768, 1986, 2871, 2034 ],
  score: 0.31280866265296936
}
{
  text: 'A picture of a camera with a blue top and green bottom.',
  bbox: [ 2112, 900, 2207, 993 ],
  score: 0.30653536319732666
}
{
  text: 'In Private logo with a blue background and a black background.',
  bbox: [ 12, 2, 189, 66 ],
  score: 0.29219281673431396
}
{
  text: 'A white background with the word business in black.',
  bbox: [ 257, 1963, 404, 2054 ],
  score: 0.2809411883354187
}
{
  text: 'a blue icon on a black background',
  bbox: [ 1822, 2080, 1881, 2151 ],
  score: 0.27423274517059326
}
{
  text: 'the wordTerms is in black letters',
  bbox: [ 2915, 1986, 3001, 2036 ],
  score: 0.26257890462875366
}
{
  text: 'A red heart with arrows pointing to the right.',
  bbox: [ 1733, 2076, 1802, 2154 ],
  score: 0.2599737048149109
}
{
  text: 'unanswerable',
  bbox: [ 2036, 898, 2157, 998 ],
  score: 0.2278139889240265
}
{
  text: 'a white background with the words settings in black',
  bbox: [ 3037, 1965, 3212, 2058 ],
  score: 0.22577747702598572
}
{
  text: 'Terms in black text on a white background.',
  bbox: [ 2898, 1962, 3018, 2056 ],
  score: 0.21895670890808105
}
{
  text: 'a circle with a q in the middle',
  bbox: [ 1033, 908, 1115, 996 ],
  score: 0.2157563865184784
}
{
  text: 'A white screen with the words "How Search Works" in black.',
  bbox: [ 437, 1962, 700, 2052 ],
  score: 0.1781202256679535
}
{
  text: 'A colorful image of a phone on a white background.',
  bbox: [ 2045, 910, 2097, 982 ],
  score: 0.16958409547805786
}
{
  text: 'a diagram of a magnifying glass in the shape of a letter',
  bbox: [ 1018, 888, 1140, 1018 ],
  score: 0.16575399041175842
}
{
  text: 'a white background with the words settings in black',
  bbox: [ 3044, 1984, 3179, 2043 ],
  score: 0.1642063558101654
}
{
  text: 'privacy written in black letters on a white background',
  bbox: [ 2748, 1959, 2886, 2054 ],
  score: 0.16381850838661194
}
{
  text: 'A picture of a colorful symbol on a white background.',
  bbox: [ 2029, 894, 2121, 1004 ],
  score: 0.1553860604763031
}
{
  text: 'A blue and white image with the words "discover the was" in the middle.',
  bbox: [ 1291, 1178, 2016, 1229 ],
  score: 0.14699873328208923
}
{
  text: 'A blurry image of the words "Action".',
  bbox: [ 1742, 1990, 1884, 2034 ],
  score: 0.14613452553749084
}
{
  text: 'A green leaf logo with the word "our" in the middle.',
  bbox: [ 1342, 1988, 1420, 2031 ],
  score: 0.14592021703720093
}
{
  text: 'A green leaf is in the corner of a square.',
  bbox: [ 1341, 1987, 1387, 2032 ],
  score: 0.14535030722618103
}
{
  text: 'A black and blue window with a key in the middle.',
  bbox: [ 2178, 2074, 2242, 2155 ],
  score: 0.12950018048286438
}
{
  text: 'A series of dots in a square shape.',
  bbox: [ 2904, 160, 2995, 253 ],
  score: 0.12734144926071167
}
{
  text: 'A black and white sign that says "our third day"',
  bbox: [ 1373, 1991, 1500, 2032 ],
  score: 0.11129298806190491
}
{
  text: 'A black background with a blue and white square in the middle.',
  bbox: [ 2336, 2074, 2410, 2156 ],
  score: 0.10925993323326111
}
{
  text: 'A blue logo with a cross on it.',
  bbox: [ 1811, 2073, 1889, 2155 ],
  score: 0.10456579923629761
}
{
  text: 'A white background with the words "Settings" in black.',
  bbox: [ 3051, 1963, 3172, 2060 ],
  score: 0.10289537906646729
}
{
  text: 'A blurry image of the words "mate action" in black.',
  bbox: [ 1654, 1990, 1881, 2033 ],
  score: 0.10276812314987183
}
{
  text: 'A sign that says "our thin" in black letters.',
  bbox: [ 1374, 1990, 1466, 2032 ],
  score: 0.10066378116607666
}
{
  text: 'A picture of the letters Gmail in black and white.',
  bbox: [ 2689, 184, 2769, 234 ],
  score: 0.09574538469314575
}
{
  text: 'a blue symbol on a white background',
  bbox: [ 1343, 804, 1918, 867 ],
  score: 0.09207239747047424
}
{
  text: 'A blurry image of the words "decade of climate action"',
  bbox: [ 1483, 1990, 1778, 2031 ],
  score: 0.09177076816558838
}
{
  text: 'A white background with the words Gmail Images written in black.',
  bbox: [ 2674, 158, 2892, 255 ],
  score: 0.0909954309463501
}
{
  text: 'a blue sign with the number 1 on it',
  bbox: [ 4, 3, 90, 66 ],
  score: 0.08966496586799622
}
{
  text: 'A black and white image with the words "our third decade of climate action" written in the middle.',
  bbox: [ 1390, 1991, 1805, 2033 ],
  score: 0.08888652920722961
}
{
  text: 'A black screen with the time 3:55 PM and the date 10/12/2024.',
  bbox: [ 3066, 2082, 3192, 2155 ],
  score: 0.08063921332359314
}
{
  text: 'a blue star in the middle of a dark room',
  bbox: [ 3061, 84, 3126, 134 ],
  score: 0.078788161277771
}
{
  text: 'A black and white image of a hole in the ground.',
  bbox: [ 2906, 173, 2979, 246 ],
  score: 0.0762108564376831
}
{
  text: 'a diagram of a magnifying glass',
  bbox: [ 996, 875, 1156, 1032 ],
  score: 0.0755038857460022
}
{
  text: 'Gmail images in black and white letters on a white background',
  bbox: [ 2684, 183, 2880, 238 ],
  score: 0.06368041038513184
}
{
  text: 'A colorful circle with a blue circle in the middle.',
  bbox: [ 1223, 1173, 1282, 1235 ],
  score: 0.06355506181716919
}
{
  text: 'unanswerable',
  bbox: [ 3167, 79, 3236, 140 ],
  score: 0.06286481022834778
}
{
  text: 'A blurry image of the words "all images"',
  bbox: [ 2737, 180, 2885, 242 ],
  score: 0.06161123514175415
}
{
  text: 'The word images is in a blurry image.',
  bbox: [ 2779, 185, 2870, 235 ],
  score: 0.060730189085006714
}
{
  text: 'A picture of a camera in the shape of a circle.',
  bbox: [ 2096, 886, 2209, 1004 ],
  score: 0.05982670187950134
}
{
  text: 'A colorful banner that reads "Discover all the ways Chrome keeps you safe while you bouse"',
  bbox: [ 1206, 1168, 2053, 1249 ],
  score: 0.053081214427948
}
{
  text: 'A colorful Google logo is shown against a white background.',
  bbox: [ 1339, 646, 1902, 853 ],
  score: 0.05053266882896423
}