{"id":35446,"date":"2025-09-30T10:48:48","date_gmt":"2025-09-30T01:48:48","guid":{"rendered":"https:\/\/uxdaystokyo.com\/articles\/?post_type=glossary&#038;p=35446"},"modified":"2025-09-30T10:48:48","modified_gmt":"2025-09-30T01:48:48","slug":"vlm","status":"publish","type":"glossary","link":"https:\/\/uxdaystokyo.com\/articles\/glossary\/vlm\/","title":{"rendered":"VLM"},"content":{"rendered":"<p data-start=\"175\" data-end=\"333\">VLM\uff08Vision-Language Models\uff09\u3068\u306f\u3001\u8996\u899a\u60c5\u5831\uff08\u753b\u50cf\u30fb\u6620\u50cf\uff09\u3068\u81ea\u7136\u8a00\u8a9e\uff08\u30c6\u30ad\u30b9\u30c8\uff09\u3092\u7d71\u5408\u7684\u306b\u7406\u89e3\u3057\u51e6\u7406\u3059\u308bAI\u30e2\u30c7\u30eb\u306e\u7dcf\u79f0\u3067\u3042\u308b\u3002<\/p>\n<p data-start=\"57\" data-end=\"110\">\u753b\u50cf\u8a8d\u8b58\uff08Computer Vision\uff09 \u306e\u6280\u8853\uff08\u4f8b\uff1a\u753b\u50cf\u306e\u7279\u5fb4\u62bd\u51fa\u3001\u7269\u4f53\u691c\u51fa\u3001\u30b7\u30fc\u30f3\u7406\u89e3\uff09\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb\uff08LLM\uff09 \u306e\u6280\u8853\uff08\u4f8b\uff1a\u81ea\u7136\u8a00\u8a9e\u306e\u7406\u89e3\u3068\u751f\u6210\u3001\u6587\u8108\u63a8\u8ad6\uff09\u3092\u7d44\u307f\u5408\u308f\u305b\u3066\u69cb\u7bc9\u3055\u308c\u305fAI\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002\u4e21\u8005\u3092\u7d50\u5408\u3059\u308b\u3053\u3068\u306b\u3088\u308a\u3001\u753b\u50cf\u3068\u8a00\u8a9e\u306e\u76f8\u4e92\u95a2\u4fc2\u3092\u7406\u89e3\u3059\u308b\u3053\u3068\u304c\u53ef\u80fd\u3068\u306a\u308b\u3002<\/p>\n<p data-start=\"335\" data-end=\"443\">\u4f8b\u3048\u3070\u3001\u753b\u50cf\u306b\u6620\u3063\u3066\u3044\u308b\u3082\u306e\u3092\u30c6\u30ad\u30b9\u30c8\u3067\u8aac\u660e\u3057\u305f\u308a\u3001\u6587\u7ae0\u304b\u3089\u5bfe\u5fdc\u3059\u308b\u753b\u50cf\u3092\u751f\u6210\u3057\u305f\u308a\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u308b\u3002\u4ee3\u8868\u4f8b\u3068\u3057\u3066 <strong data-start=\"393\" data-end=\"409\"><a href=\"https:\/\/openai.com\/ja-JP\/index\/clip\/\" target=\"_blank\" rel=\"noopener\">CLIP<\/a>\uff08OpenAI\uff09<\/strong> \u3084 <strong data-start=\"412\" data-end=\"434\">Flamingo\uff08<a href=\"https:\/\/deepmind.google\/?_gl=1*tznbxz*_up*MQ..*_ga*MTMxMTEzNjU4LjE3NTkxOTQ1Nzk.*_ga_LS8HVHCNQ0*czE3NTkxOTQ1NzgkbzEkZzAkdDE3NTkxOTQ1NzgkajYwJGwwJGgw\">DeepMind<\/a>\uff09<\/strong> \u306a\u3069\u304c\u3042\u308b\u3002DeepMind \u306f 2014\u5e74\u306b Google \u306b\u8cb7\u53ce\u3055\u308c\u3001\u305d\u306e\u5f8c\u300cGoogle DeepMind\u300d\u3068\u3044\u3046\u540d\u79f0\u3067\u7814\u7a76\u3092\u7d9a\u3051\u3066\u3044\u308b\u3002<\/p>\n<p data-start=\"101\" data-end=\"159\">\u300c\u30d6\u30a4\u30fb\u30a8\u30eb\u30fb\u30a8\u30e0\u300d\u3068\u8aad\u3080\u306e\u304c\u4e00\u822c\u7684\u3067\u3042\u308b\u3002\u65e5\u672c\u8a9e\u8868\u8a18\u3067\u306f\u300c\u30d3\u30b8\u30e7\u30f3\u30fb\u30e9\u30f3\u30b2\u30fc\u30b8\u30fb\u30e2\u30c7\u30eb\u300d\u3068\u8868\u73fe\u3059\u308b\u3053\u3068\u3082\u3042\u308b\u3002\u307e\u305f\u3001<span>\u65e5\u672c\u8a9e\u3067\u306f\u300c\u5927\u898f\u6a21\u8996\u899a\u8a00\u8a9e\u30e2\u30c7\u30eb\u300d\u3068\u547c\u3070\u308c\u308b\u3002<\/span><\/p>\n<h2 data-start=\"101\" data-end=\"159\">\u51fa\u6765\u308b\u3053\u3068<\/h2>\n<ul>\n<li data-uid=\"hPiLviY2\" data-time=\"1733190928419\">\u753b\u50cf\u3092\u8aac\u660e\u3059\u308b\u6587\u7ae0\u3092\u751f\u6210\u3059\u308b\u3001<strong>\u753b\u50cf\u30ad\u30e3\u30d7\u30b7\u30e7\u30f3\u751f\u6210<\/strong><\/li>\n<li data-uid=\"SgRqOK5r\" data-time=\"1733190928419\">\u300c\u3053\u306e\u753b\u50cf\u306b\u4f55\u304c\u5199\u3063\u3066\u3044\u308b\u304b\u6559\u3048\u3066\u300d\u3068\u3044\u3063\u305f<strong>\u753b\u50cf\u306b\u57fa\u3065\u304f\u8cea\u554f\u5fdc\u7b54<\/strong><\/li>\n<li data-uid=\"X1AT6bAf\" data-time=\"1733190928419\">\u30c6\u30ad\u30b9\u30c8\u5165\u529b\u304b\u3089\u753b\u50cf\u306e\u53d6\u5f97\u3001\u307e\u305f\u306f\u753b\u50cf\u5165\u529b\u304b\u3089\u30c6\u30ad\u30b9\u30c8\u306b\u3088\u308b\u8aac\u660e\u306e\u53d6\u5f97\u3001\u3068\u3044\u3063\u305f\u7570\u306a\u308b\u30c7\u30fc\u30bf\u5f62\u5f0f\u3092\u7d44\u307f\u5408\u308f\u305b\u305f<strong>\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u691c\u7d22<\/strong><\/li>\n<\/ul>\n<h2 data-start=\"450\" data-end=\"456\">\u63d0\u5531\u8005<\/h2>\n<div style=\"width: 363px\" class=\"wp-caption alignnone\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/www.czbiohub.org\/wp-content\/uploads\/2024\/03\/sam-altman.png?w=761\" width=\"353\" height=\"353\" alt=\"\u30b5\u30e0\u30fb\u30a2\u30eb\u30c8\u30de\u30f3OpenAI\u6700\u9ad8\u7d4c\u55b6\u8cac\u4efb\u8005\" class=\"\" \/><p class=\"wp-caption-text\"><a href=\"https:\/\/www.czbiohub.org\/person\/sam-altman\/\">\u30b5\u30e0\u30fb\u30a2\u30eb\u30c8\u30de\u30f3<\/a> \u6c0f OpenAI\u6700\u9ad8\u7d4c\u55b6\u8cac\u4efb\u8005<\/p><\/div>\n<div style=\"width: 366px\" class=\"wp-caption alignnone\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/9\/94\/Demis_Hassabis_in_2025_by_Christopher_Michel_A.jpg\" width=\"356\" height=\"534\" alt=\"DeepMind \u5171\u540c\u5275\u696d\u8005 \u30c7\u30df\u30b9\u30fb\u30cf\u30b5\u30d3\u30b9\u6c0f\uff08Demis Hassabis\uff09\" \/><p class=\"wp-caption-text\">DeepMind \u5171\u540c\u5275\u696d\u8005 <a href=\"https:\/\/en.wikipedia.org\/wiki\/Demis_Hassabis\">\u30c7\u30df\u30b9\u30fb\u30cf\u30b5\u30d3\u30b9<\/a>\u6c0f\uff08Demis Hassabis\uff09<\/p><\/div>\n<p data-start=\"457\" data-end=\"571\">\u300cVLM\u300d\u3068\u3044\u3046\u6982\u5ff5\u3092\u5358\u72ec\u3067\u63d0\u5531\u3057\u305f\u7279\u5b9a\u306e\u7814\u7a76\u8005\u306f\u3044\u306a\u3044\u3002\u305f\u3060\u3057\u3001OpenAI\uff08CLIP, 2021\u5e74\uff09 \u3084 DeepMind\uff08Flamingo, 2022\u5e74\uff09 \u306a\u3069\u306e\u7814\u7a76\u304c\u3001\u3053\u306e\u5206\u91ce\u3092\u5927\u304d\u304f\u63a8\u9032\u3057\u305f\u3002<\/p>\n<h2 data-uid=\"i9cYhgt_\" data-time=\"1733190983341\" id=\"index_i9cYhgt_\">VLM\u306e\u6d3b\u7528\u4e8b\u4f8b<\/h2>\n<ul>\n<li data-uid=\"JmCjnO6y\" data-time=\"1733190989557\">\u81ea\u52d5\u904b\u8ee2\u3067\u306e<span>\u4ea4\u901a\u6a19\u8b58\u3084\u9053\u8def\u6307\u793a<\/span><\/li>\n<li data-uid=\"JmCjnO6y\" data-time=\"1733190989557\">EC\u30b5\u30a4\u30c8\u3067\u306e\u5546\u54c1\u30ec\u30b3\u30e1\u30f3\u30c9<\/li>\n<li data-uid=\"JmCjnO6y\" data-time=\"1733190989557\">\u5546\u54c1\u306a\u3069\u306e\u60c5\u5831\u306e\u81ea\u52d5\u751f\u6210<\/li>\n<li data-uid=\"JmCjnO6y\" data-time=\"1733190989557\"><\/li>\n<\/ul>\n<h2 data-start=\"921\" data-end=\"943\">\u30c7\u30b6\u30a4\u30f3\u4e0a\u306b\u95a2\u308f\u308b\u5229\u7528\u65b9\u6cd5\u3068\u5177\u4f53\u7684\u4e8b\u4f8b<\/h2>\n<h3 data-start=\"947\" data-end=\"963\"><strong data-start=\"947\" data-end=\"961\">\u30a2\u30af\u30bb\u30b7\u30d3\u30ea\u30c6\u30a3\u5411\u4e0a<\/strong><\/h3>\n<ul>\n<li data-start=\"969\" data-end=\"1010\">VLM\u306f\u8996\u899a\u969c\u5bb3\u8005\u5411\u3051\u306b\u3001\u753b\u50cf\u3084UI\u3092\u81ea\u7136\u8a00\u8a9e\u3067\u8aac\u660e\u3059\u308b\u30b7\u30b9\u30c6\u30e0\u306b\u6d3b\u7528\u3067\u304d\u308b\u3002<\/li>\n<li data-start=\"1016\" data-end=\"1057\">\u4f8b\uff1a\u5199\u771f\u3092\u30a2\u30c3\u30d7\u30ed\u30fc\u30c9\u3059\u308b\u3068\u300c3\u4eba\u304c\u4f1a\u8b70\u5ba4\u3067\u8b70\u8ad6\u3057\u3066\u3044\u308b\u69d8\u5b50\u3067\u3059\u300d\u3068\u8aac\u660e\u3059\u308b\u3002<\/li>\n<\/ul>\n<h3 data-start=\"1062\" data-end=\"1075\"><strong data-start=\"1062\" data-end=\"1073\">\u691c\u7d22\u4f53\u9a13\u306e\u6539\u5584<\/strong><\/h3>\n<ul>\n<li data-start=\"1081\" data-end=\"1135\">\u30e6\u30fc\u30b6\u30fc\u304c\u300c\u8d64\u3044\u82b1\u306e\u54b2\u3044\u305f\u5ead\u306e\u5199\u771f\u300d\u3068\u5165\u529b\u3059\u308b\u3068\u3001\u81a8\u5927\u306a\u753b\u50cf\u30c7\u30fc\u30bf\u30d9\u30fc\u30b9\u304b\u3089\u8a72\u5f53\u753b\u50cf\u3092\u5373\u5ea7\u306b\u691c\u7d22\u3067\u304d\u308b\u3002<\/li>\n<\/ul>\n<h3 data-start=\"1140\" data-end=\"1156\"><strong data-start=\"1140\" data-end=\"1154\">\u30c7\u30b6\u30a4\u30f3\u30ea\u30b5\u30fc\u30c1\u652f\u63f4<\/strong><\/h3>\n<ul>\n<li data-start=\"1162\" data-end=\"1220\">\u30c7\u30b6\u30a4\u30ca\u30fc\u304c\u300c\u30df\u30cb\u30de\u30eb\u3067\u5317\u6b27\u98a8\u306e\u5bb6\u5177\u30c7\u30b6\u30a4\u30f3\u300d\u3068\u5165\u529b\u3059\u308b\u3068\u3001\u95a2\u9023\u3059\u308b\u753b\u50cf\u3092\u63d0\u793a\u3057\u3001\u30a4\u30f3\u30b9\u30d4\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u5f97\u3089\u308c\u308b\u3002<\/li>\n<\/ul>\n<h3 data-start=\"1225\" data-end=\"1243\"><strong data-start=\"1225\" data-end=\"1241\">\u30d7\u30ed\u30c8\u30bf\u30a4\u30d4\u30f3\u30b0\u306e\u52b9\u7387\u5316<\/strong><\/h3>\n<ul>\n<li data-start=\"1249\" data-end=\"1301\">\u30b9\u30b1\u30c3\u30c1\u3084UI\u306e\u30ef\u30a4\u30e4\u30fc\u30d5\u30ec\u30fc\u30e0\u3092\u8aad\u307f\u53d6\u308a\u3001\u300c\u3053\u306e\u753b\u9762\u306f\u30ed\u30b0\u30a4\u30f3\u30da\u30fc\u30b8\u3067\u3059\u300d\u3068\u81ea\u52d5\u3067\u30c6\u30ad\u30b9\u30c8\u5316\u3059\u308b\u3002<\/li>\n<li data-start=\"1307\" data-end=\"1330\">\u30c7\u30b6\u30a4\u30f3\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u751f\u6210\u3092\u534a\u81ea\u52d5\u5316\u3067\u304d\u308b\u3002<\/li>\n<\/ul>\n<h2 data-start=\"1337\" data-end=\"1367\">\u30d7\u30ed\u30c0\u30af\u30c8\u30fb\u30b3\u30f3\u30c6\u30f3\u30c4\u30c7\u30b6\u30a4\u30f3\u306e\u89b3\u70b9\u3067\u4f7f\u3048\u308b\u5834\u9762\u3068\u4e8b\u4f8b<\/h2>\n<h3 data-start=\"1370\" data-end=\"1450\"><strong data-start=\"1370\" data-end=\"1377\">\u5834\u9762\u2460<\/strong> \u65b0\u898f\u30a2\u30d7\u30ea\u306eUI\u30c6\u30b9\u30c8<\/h3>\n<p data-start=\"1370\" data-end=\"1450\">\u2192 \u624b\u63cf\u304d\u306e\u753b\u9762\u8a2d\u8a08\u56f3\u3092\u5199\u771f\u306b\u64ae\u308b\u3068\u3001VLM\u304c\u300c\u30db\u30fc\u30e0\u753b\u9762\u300d\u300c\u30d7\u30ed\u30d5\u30a3\u30fc\u30eb\u753b\u9762\u300d\u306a\u3069\u3092\u5206\u985e\u30fb\u8aac\u660e\u3057\u3066\u304f\u308c\u308b\u3002<\/p>\n<h3 data-start=\"1454\" data-end=\"1537\"><strong data-start=\"1454\" data-end=\"1461\">\u5834\u9762\u2461<\/strong> E\u30b3\u30de\u30fc\u30b9\u306b\u304a\u3051\u308b\u5546\u54c1\u691c\u7d22<\/h3>\n<p data-start=\"1454\" data-end=\"1537\">\u2192 \u30e6\u30fc\u30b6\u30fc\u304c\u300c\u9752\u3044\u30c1\u30a7\u30c3\u30af\u67c4\u306e\u30b7\u30e3\u30c4\u3092\u63a2\u3057\u3066\u3044\u308b\u300d\u3068\u5165\u529b\u3059\u308b\u3068\u3001\u5546\u54c1\u753b\u50cf\u3092\u8a00\u8a9e\u7684\u306b\u7406\u89e3\u3057\u3066\u30de\u30c3\u30c1\u30f3\u30b0\u3067\u304d\u308b\u3002<\/p>\n<h3 data-start=\"1541\" data-end=\"1625\"><strong data-start=\"1541\" data-end=\"1548\">\u5834\u9762\u2462<\/strong> \u6559\u80b2\u3084\u5b66\u7fd2\u652f\u63f4<\/h3>\n<p data-start=\"1541\" data-end=\"1625\">\u2192 \u6559\u6750\u306e\u753b\u50cf\uff08\u4f8b\uff1a\u5730\u56f3\u3084\u30b0\u30e9\u30d5\uff09\u3092\u30a2\u30c3\u30d7\u30ed\u30fc\u30c9\u3059\u308b\u3068\u3001VLM\u304c\u300c\u3053\u308c\u306f\u7b2c\u4e8c\u6b21\u4e16\u754c\u5927\u6226\u306e\u6226\u6cc1\u3092\u793a\u3057\u305f\u5730\u56f3\u3067\u3059\u300d\u3068\u89e3\u8aac\u3067\u304d\u308b\u3002<\/p>\n<p>&nbsp;<\/p>\n<h2>6\u7a2e\u985e\u306e\u751f\u6210AI\u30e2\u30c7\u30eb<\/h2>\n<p>\u751f\u6210AI\u5168\u4f53\u3068\u3057\u3066\u306f \u8a00\u8a9e\u30fb\u753b\u50cf\u30fb\u97f3\u58f0\u30fb\u52d5\u753b\u30fb3D\u30fb\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb \u306a\u3069\u591a\u69d8\u306a\u5206\u91ce\u3067\u6d3b\u7528\u3055\u308c\u3066\u3044\u308b\u3002<\/p>\n<h3 data-start=\"92\" data-end=\"104\">1.\u8a00\u8a9e\u30e2\u30c7\u30eb<\/h3>\n<ul>\n<li><a href=\"https:\/\/uxdaystokyo.com\/articles\/glossary\/large-language-model\/\"><strong data-start=\"107\" data-end=\"147\">LLM<span>\uff08Large Language Model\uff09<\/span><\/strong><\/a><br data-start=\"147\" data-end=\"150\" \/>2017\u5e74\u306bGoogle\u306e\u7814\u7a76\u30c1\u30fc\u30e0\u304c\u767a\u8868\u3057\u305f\u300cAttention is All You Need<strong data-start=\"405\" data-end=\"459\">\u300d<\/strong><span>\u00a0\u8ad6\u6587<\/span><br data-start=\"250\" data-end=\"253\" \/><strong data-start=\"255\" data-end=\"261\">\u4e8b\u4f8b<\/strong>\uff1a\u30ab\u30b9\u30bf\u30de\u30fc\u30b5\u30dd\u30fc\u30c8\u3084\u30ec\u30b9\u30c8\u30e9\u30f3\u4e88\u7d04\u306a\u3069\u3002<\/li>\n<\/ul>\n<h3 data-start=\"455\" data-end=\"470\">2.\u753b\u50cf\u751f\u6210\u7cfb<\/h3>\n<ul data-start=\"105\" data-end=\"448\">\n<li data-start=\"105\" data-end=\"296\">\n<p data-start=\"107\" data-end=\"296\"><a href=\"https:\/\/uxdaystokyo.com\/articles\/glossary\/gan\/\"><strong data-start=\"107\" data-end=\"147\">GAN\uff08Generative Adversarial Networks\uff09<\/strong><\/a><br data-start=\"147\" data-end=\"150\" \/>\u751f\u6210\u5668\uff08Generator\uff09\u3068\u8b58\u5225\u5668\uff08Discriminator\uff09\u306e\u4e8c\u8005\u304c\u7af6\u3044\u5408\u3046\u4ed5\u7d44\u307f\u3092\u5229\u7528\u3057\u3001\u30ea\u30a2\u30eb\u306a\u753b\u50cf\u3084\u6620\u50cf\u3092\u751f\u6210\u3059\u308b\u3002<br data-start=\"250\" data-end=\"253\" \/><strong data-start=\"255\" data-end=\"261\">\u4e8b\u4f8b<\/strong>\uff1a\u4eba\u7269\u306e\u30d5\u30a7\u30a4\u30af\u5199\u771f\u751f\u6210\u3001\u30d5\u30a1\u30c3\u30b7\u30e7\u30f3\u30c7\u30b6\u30a4\u30f3\u306e\u30d7\u30ed\u30c8\u30bf\u30a4\u30d7\u4f5c\u6210\u3002<\/p>\n<\/li>\n<li data-start=\"298\" data-end=\"448\">\n<p data-start=\"300\" data-end=\"448\"><strong data-start=\"300\" data-end=\"356\">\u62e1\u6563\u30e2\u30c7\u30eb\uff08Diffusion Models, \u4f8b: Stable Diffusion, DALL\u00b7E\uff09<\/strong><br data-start=\"356\" data-end=\"359\" \/>\u30ce\u30a4\u30ba\u304b\u3089\u5f90\u3005\u306b\u753b\u50cf\u3092\u5fa9\u5143\u3059\u308b\u4ed5\u7d44\u307f\u3092\u6301\u3064\u3002<a href=\"https:\/\/uxdaystokyo.com\/articles\/glossary\/gan\/\">GAN<\/a>\u3088\u308a\u5b89\u5b9a\u7684\u3067\u9ad8\u89e3\u50cf\u5ea6\u306e\u751f\u6210\u304c\u53ef\u80fd\u3067\u3042\u308b\u3002<br data-start=\"406\" data-end=\"409\" \/><strong data-start=\"411\" data-end=\"417\">\u4e8b\u4f8b<\/strong>\uff1a\u5e83\u544a\u30d3\u30b8\u30e5\u30a2\u30eb\u306e\u4f5c\u6210\u3001\u30d7\u30ed\u30c0\u30af\u30c8\u306e\u30b3\u30f3\u30bb\u30d7\u30c8\u30a2\u30fc\u30c8\u751f\u6210\u3002<\/p>\n<\/li>\n<\/ul>\n<ul data-start=\"471\" data-end=\"726\"><\/ul>\n<h3 data-start=\"733\" data-end=\"745\">3. \u52d5\u753b\u751f\u6210\u7cfb<\/h3>\n<ul data-start=\"746\" data-end=\"878\">\n<li data-start=\"746\" data-end=\"878\">\n<p data-start=\"748\" data-end=\"878\"><strong data-start=\"748\" data-end=\"778\">Video Diffusion \/ VideoGAN<\/strong><br data-start=\"778\" data-end=\"781\" \/>\u62e1\u6563\u30e2\u30c7\u30eb\u3084<a href=\"https:\/\/uxdaystokyo.com\/articles\/glossary\/gan\/\">GAN<\/a>\u3092\u52d5\u753b\u306b\u62e1\u5f35\u3057\u305f\u3082\u306e\u3067\u3042\u308b\u3002\u9023\u7d9a\u3059\u308b\u30d5\u30ec\u30fc\u30e0\u3092\u81ea\u7136\u306b\u3064\u306a\u304e\u3001\u9ad8\u54c1\u8cea\u306a\u6620\u50cf\u3092\u751f\u6210\u3067\u304d\u308b\u3002<br data-start=\"835\" data-end=\"838\" \/><strong data-start=\"840\" data-end=\"846\">\u4e8b\u4f8b<\/strong>\uff1aUI\u30a2\u30cb\u30e1\u30fc\u30b7\u30e7\u30f3\u306e\u30d7\u30ed\u30c8\u30bf\u30a4\u30d7\u4f5c\u6210\u3001\u5e83\u544a\u52d5\u753b\u306e\u81ea\u52d5\u751f\u6210\u3002<\/p>\n<\/li>\n<\/ul>\n<h3 data-start=\"885\" data-end=\"897\">4.\u97f3\u58f0\u30fb\u97f3\u697d\u751f\u6210\u7cfb<\/h3>\n<ul data-start=\"471\" data-end=\"726\">\n<li data-start=\"471\" data-end=\"612\">\n<p data-start=\"473\" data-end=\"612\"><strong data-start=\"473\" data-end=\"526\">\u97f3\u58f0\u5408\u6210\u30e2\u30c7\u30eb\uff08TTS: Text-to-Speech, \u4f8b: Tacotron, VALL-E\uff09<\/strong><br data-start=\"526\" data-end=\"529\" \/>\u30c6\u30ad\u30b9\u30c8\u3092\u4eba\u9593\u3089\u3057\u3044\u97f3\u58f0\u306b\u5909\u63db\u3059\u308b\u3002\u58f0\u8cea\u3084\u611f\u60c5\u8868\u73fe\u306e\u8abf\u6574\u3082\u53ef\u80fd\u3067\u3042\u308b\u3002<br data-start=\"566\" data-end=\"569\" \/><strong data-start=\"571\" data-end=\"577\">\u4e8b\u4f8b<\/strong>\uff1a\u30ca\u30ec\u30fc\u30b7\u30e7\u30f3\u4ed8\u304dUX\u30d7\u30ed\u30c8\u30bf\u30a4\u30d7\u3001\u30ad\u30e3\u30e9\u30af\u30bf\u30fc\u30dc\u30a4\u30b9\u306e\u81ea\u52d5\u751f\u6210\u3002<\/p>\n<\/li>\n<li data-start=\"614\" data-end=\"726\">\n<p data-start=\"616\" data-end=\"726\"><strong data-start=\"616\" data-end=\"648\">\u97f3\u697d\u751f\u6210\u30e2\u30c7\u30eb\uff08\u4f8b: Jukebox, MusicLM\uff09<\/strong><br data-start=\"648\" data-end=\"651\" \/>\u7279\u5b9a\u306e\u30b8\u30e3\u30f3\u30eb\u3084\u697d\u5668\u69cb\u6210\u306b\u57fa\u3065\u304d\u65b0\u3057\u3044\u697d\u66f2\u3092\u751f\u6210\u3059\u308b\u3002<br data-start=\"680\" data-end=\"683\" \/><strong data-start=\"685\" data-end=\"691\">\u4e8b\u4f8b<\/strong>\uff1a\u30a2\u30d7\u30ea\u5185\u306eBGM\u751f\u6210\u3001\u30e6\u30fc\u30b6\u30fc\u4f53\u9a13\u306b\u5408\u308f\u305b\u305f\u97f3\u306e\u30d1\u30fc\u30bd\u30ca\u30e9\u30a4\u30ba\u3002<\/p>\n<\/li>\n<\/ul>\n<ul data-start=\"898\" data-end=\"1115\"><\/ul>\n<h3 data-start=\"1122\" data-end=\"1139\">5. \u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30eb\u751f\u6210\u7cfb<\/h3>\n<ul data-start=\"1140\" data-end=\"1381\">\n<li data-start=\"1140\" data-end=\"1258\">\n<p data-start=\"1142\" data-end=\"1258\"><strong data-start=\"1142\" data-end=\"1177\">CLIP + Diffusion\uff08OpenAI\u306eCLIP\u306a\u3069\uff09<\/strong><br data-start=\"1177\" data-end=\"1180\" \/>\u753b\u50cf\u3068\u8a00\u8a9e\u3092\u7d44\u307f\u5408\u308f\u305b\u3001\u3088\u308a\u610f\u5473\u7684\u306b\u6574\u5408\u6027\u306e\u3042\u308b\u751f\u6210\u3092\u5b9f\u73fe\u3059\u308b\u3002<br data-start=\"1214\" data-end=\"1217\" \/><strong data-start=\"1219\" data-end=\"1225\">\u4e8b\u4f8b<\/strong>\uff1a\u30c6\u30ad\u30b9\u30c8\u304b\u3089UI\u30c7\u30b6\u30a4\u30f3\u3084\u8aac\u660e\u56f3\u3092\u751f\u6210\u3001\u6559\u80b2\u6559\u6750\u306e\u81ea\u52d5\u4f5c\u6210\u3002<\/p>\n<\/li>\n<li data-start=\"1260\" data-end=\"1381\">\n<p data-start=\"1262\" data-end=\"1381\"><strong data-start=\"1262\" data-end=\"1316\">VLM\uff08Vision-Language Models, \u4f8b: Flamingo, Kosmos-1\uff09<\/strong><br data-start=\"1316\" data-end=\"1319\" \/>\u30c6\u30ad\u30b9\u30c8\u3068\u753b\u50cf\u3092\u7d71\u5408\u7684\u306b\u7406\u89e3\u30fb\u751f\u6210\u3059\u308b\u3002<br data-start=\"1341\" data-end=\"1344\" \/><strong data-start=\"1346\" data-end=\"1352\">\u4e8b\u4f8b<\/strong>\uff1a\u30e6\u30fc\u30b6\u30fc\u304c\u63cf\u3044\u305f\u30b9\u30b1\u30c3\u30c1\u3092\u81ea\u52d5\u7684\u306b\u30c7\u30b6\u30a4\u30f3\u6848\u306b\u5909\u63db\u3002<\/p>\n<\/li>\n<\/ul>\n<h3 data-start=\"1122\" data-end=\"1139\">6.3D\u751f\u6210\u7cfb<\/h3>\n<ul data-start=\"898\" data-end=\"1115\">\n<li data-start=\"898\" data-end=\"1015\">\n<p data-start=\"900\" data-end=\"1015\"><strong data-start=\"900\" data-end=\"932\">NeRF\uff08Neural Radiance Fields\uff09<\/strong><br data-start=\"932\" data-end=\"935\" \/>\u8907\u6570\u679a\u306e\u5199\u771f\u304b\u30893D\u30e2\u30c7\u30eb\u3092\u751f\u6210\u3059\u308b\u30023D\u7a7a\u9593\u3067\u306e\u30ec\u30f3\u30c0\u30ea\u30f3\u30b0\u306b\u5f37\u3044\u3002<br data-start=\"972\" data-end=\"975\" \/><strong data-start=\"977\" data-end=\"983\">\u4e8b\u4f8b<\/strong>\uff1a\u88fd\u54c1\u30c7\u30b6\u30a4\u30f3\u306e\u30e2\u30c3\u30af\u30a2\u30c3\u30d7\u3001\u30b2\u30fc\u30e0\u75283D\u30aa\u30d6\u30b8\u30a7\u30af\u30c8\u751f\u6210\u3002<\/p>\n<\/li>\n<li data-start=\"1017\" data-end=\"1115\">\n<p data-start=\"1019\" data-end=\"1115\"><strong data-start=\"1019\" data-end=\"1046\">3D\u62e1\u6563\u30e2\u30c7\u30eb\uff08\u4f8b: DreamFusion\uff09<\/strong><br data-start=\"1046\" data-end=\"1049\" \/>\u30c6\u30ad\u30b9\u30c8\u304b\u3089\u76f4\u63a53D\u30e2\u30c7\u30eb\u3092\u751f\u6210\u3067\u304d\u308b\u3002<br data-start=\"1071\" data-end=\"1074\" \/><strong data-start=\"1076\" data-end=\"1082\">\u4e8b\u4f8b<\/strong>\uff1a\u5bb6\u5177\u3084\u5bb6\u96fb\u306e\u30b3\u30f3\u30bb\u30d7\u30c8\u30c7\u30b6\u30a4\u30f3\u3001\u5efa\u7bc9\u30d3\u30b8\u30e5\u30a2\u30e9\u30a4\u30bc\u30fc\u30b7\u30e7\u30f3\u3002<\/p>\n<\/li>\n<\/ul>\n<h2>\u95a2\u9023\u7528\u8a9e<\/h2>\n<ul>\n<li><a href=\"https:\/\/uxdaystokyo.com\/articles\/glossary\/generative-ai\/\">\u751f\u6210AI<\/a><\/li>\n<li><a href=\"https:\/\/uxdaystokyo.com\/articles\/glossary\/ai-agent\/\">AI\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>\u753b\u50cf\u30fb\u6620\u50cf\u3068\u81ea\u7136\u8a00\u8a9e\u3092\u7d71\u5408\u7684\u306b\u7406\u89e3\u3057\u51e6\u7406\u3059\u308bAI\u30e2\u30c7\u30eb<\/p>\n","protected":false},"author":5,"featured_media":35450,"template":"","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"English":"Vision-Language Models","footnotes":""},"glossary-category":[418],"glossary_tag":[],"class_list":["post-35446","glossary","type-glossary","status-publish","has-post-thumbnail","hentry","glossary-category-technology"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/glossary\/35446","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/glossary"}],"about":[{"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/types\/glossary"}],"author":[{"embeddable":true,"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/users\/5"}],"version-history":[{"count":2,"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/glossary\/35446\/revisions"}],"predecessor-version":[{"id":35451,"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/glossary\/35446\/revisions\/35451"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/media\/35450"}],"wp:attachment":[{"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/media?parent=35446"}],"wp:term":[{"taxonomy":"glossary-category","embeddable":true,"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/glossary-category?post=35446"},{"taxonomy":"glossary_tag","embeddable":true,"href":"https:\/\/uxdaystokyo.com\/articles\/wp-json\/wp\/v2\/glossary_tag?post=35446"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}