{"id":8939,"date":"2026-06-09T00:04:31","date_gmt":"2026-06-08T16:04:31","guid":{"rendered":"https:\/\/infernews.com\/blog\/official-code-repository-for-quot-stream3d-vlm-online-3d-spatial-understanding-w\/"},"modified":"2026-06-09T00:09:25","modified_gmt":"2026-06-08T16:09:25","slug":"official-code-repository-for-quot-stream3d-vlm-online-3d-spatial-understanding-w","status":"publish","type":"post","link":"https:\/\/infernews.com\/blog\/official-code-repository-for-quot-stream3d-vlm-online-3d-spatial-understanding-w\/","title":{"rendered":"Stream3D-VLM \u628a\u4e32\u6d41\u5f71\u7247\u8b8a\u6210 3D"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/06\/logo-9e3bd55001e0.jpg\" alt=\"Stream3D-VLM Logo\"\/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Stream3D-VLM \u662f\u4e00\u500b online 3D vision-language model\uff0c\u91cd\u9ede\u662f\u76f4\u63a5\u5f9e\u4e32\u6d41\u5f71\u7247\u505a\u5373\u6642\u7a7a\u9593\u7406\u89e3\uff0c\u800c\u4e0d\u662f\u7b49\u6574\u6bb5\u5f71\u7247\u6216\u6574\u500b\u5834\u666f\u6536\u96c6\u5b8c\u624d\u5206\u6790\u3002\u5c0d\u60f3\u7814\u7a76\u6a5f\u68b0\u4eba\u3001\u7a7a\u9593\u554f\u7b54\uff0c\u6216 3D \u5834\u666f\u4e92\u52d5\u7684\u4eba\u4f86\u8aaa\uff0c\u9019\u500b\u9805\u76ee\u8655\u7406\u7684\u662f\u300c\u6a21\u578b\u53ef\u5426\u4e00\u908a\u770b\u3001\u4e00\u908a\u5efa\u7acb\u5834\u666f\u6982\u5ff5\uff0c\u518d\u5373\u6642\u56de\u7b54\u554f\u984c\u300d\u9019\u4ef6\u4e8b\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u4f7f\u7528\u9019\u500b\u9805\u76ee\u6642\uff0c\u6838\u5fc3\u8cc7\u6e90\u5305\u62ec\u5df2\u516c\u958b\u7684 Stream3D-VLM-4B \u6a21\u578b\u3001Stream3D-1M Dataset\uff0c\u4ee5\u53ca Stream3D-Bench\u3002\u8cc7\u6599\u65b9\u9762\u672a\u6709\u76f4\u63a5\u91cb\u51fa\u539f\u59cb\u5a92\u9ad4\uff0c\u4f46\u6709\u63d0\u4f9b\u6a19\u8a3b\u3001GLB \u8207 RRD \u7b49\u91cd\u5efa\u7d50\u679c\uff1bGLB \u53ef\u653e\u5165\u4e00\u822c 3D viewer \u9010\u6b65\u67e5\u770b\u9ede\u96f2\uff0cRRD \u5247\u53ef\u914d\u5408\u76f8\u6a5f\u59ff\u614b\u8207\u9ede\u96f2\u8a18\u9304\u89c0\u5bdf\u5b8c\u6574\u91cd\u5efa\u6d41\u7a0b\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Stream3D-VLM \u52a0\u5165 incremental geometry priors\uff0c\u4ee4\u6a21\u578b\u96a8\u6642\u9593\u5438\u6536\u5c0d\u9f4a\u7684 3D \u5e7e\u4f55\u7dda\u7d22\u3002\u9805\u76ee\u4ea6\u63d0\u51fa Visual-Spatial Feature Integration\uff08VSFI\uff09\u6a21\u7d44\uff0c\u4ee5\u53ca Geometry-Adaptive Voxel Compression\uff08GAVC\uff09\u6a21\u7d44\uff0c\u524d\u8005\u8ca0\u8cac\u628a\u5e7e\u4f55\u8cc7\u8a0a\u9010\u6b65\u6ce8\u5165\u8996\u89ba\u4e32\u6d41\uff0c\u5f8c\u8005\u7528 3D \u7d50\u69cb\u53bb\u58d3\u7e2e visual tokens\uff0c\u6e1b\u5c11\u9577\u5e8f\u5217\u63a8\u7406\u8ca0\u64d4\u3002<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><img data-dominant-color=\"d1d0ce\" data-has-transparency=\"false\" style=\"--dominant-color: #d1d0ce;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"531\" src=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/06\/image-1.png\" alt=\"\" class=\"wp-image-8940 not-transparent\" srcset=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/06\/image-1.png 1024w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/06\/image-1-300x156.png 300w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/06\/image-1-768x398.png 768w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">\u5718\u968a\u4ea6\u5efa\u7acb\u4e86\u53ef\u64f4\u5c55\u7684\u8cc7\u6599\u751f\u6210\u6d41\u7a0b\uff0c\u6574\u7406\u8d85\u904e 1M online spatio-temporal 3D QA pairs\uff0c\u4e26\u8a2d\u8a08\u6db5\u84cb 29 \u9805\u4efb\u52d9\u7684\u57fa\u6e96\u3002\u9805\u76ee\u8072\u7a31\u5728 online \u8207 offline \u7684 3D spatial understanding\u3001reasoning\u3001grounding \u4efb\u52d9\u4e0a\uff0c\u8868\u73fe\u512a\u65bc\u90e8\u5206 proprietary \u8207 open-source models\uff1b\u4e0d\u904e\u6587\u7ae0\u672a\u5728\u9019\u4efd\u8cc7\u8a0a\u4e2d\u5217\u51fa\u5b8c\u6574\u6578\u5b57\uff0c\u95b1\u8b80\u7d50\u679c\u6642\u4ecd\u8981\u914d\u5408\u8ad6\u6587\u8207\u5be6\u9a57\u9801\u9762\u4e00\u8d77\u770b\u3002<\/p>\n\n\n<div class=\"align wp-block-vpb-video\" id='vpbVideoPlayer-1' data-attributes='{&quot;source&quot;:&quot;https:\\\/\\\/stream3d-vlm.github.io\\\/images\\\/project_video.mp4&quot;,&quot;repeat&quot;:true,&quot;autoplay&quot;:true,&quot;muted&quot;:true,&quot;align&quot;:&quot;&quot;,&quot;poster&quot;:&quot;&quot;,&quot;controls&quot;:{&quot;play-large&quot;:true,&quot;restart&quot;:false,&quot;rewind&quot;:true,&quot;play&quot;:true,&quot;fast-forward&quot;:true,&quot;progress&quot;:true,&quot;current-time&quot;:true,&quot;duration&quot;:false,&quot;mute&quot;:true,&quot;volume&quot;:true,&quot;pip&quot;:false,&quot;airplay&quot;:false,&quot;settings&quot;:true,&quot;download&quot;:false,&quot;fullscreen&quot;:true},&quot;width&quot;:&quot;100%&quot;,&quot;radius&quot;:&quot;0px&quot;,&quot;resetOnEnd&quot;:false,&quot;autoHideControl&quot;:true}'><\/div>\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u89e3\u6c7a\u75db\u9ede<\/strong>\uff1a\u50b3\u7d71 3D Large Multimodal Models \u591a\u6578\u4f9d\u8cf4\u96e2\u7dda\u8655\u7406\uff0c\u9019\u500b\u9805\u76ee\u6539\u70ba\u9762\u5411 streaming video\u3002<\/li>\n\n\n\n<li><strong>\u4e3b\u8981\u65b9\u6cd5<\/strong>\uff1a\u7d50\u5408 autoregressive streaming control\u3001VSFI \u8207 GAVC\u3002<\/li>\n\n\n\n<li><strong>\u8cc7\u6599\u8207\u57fa\u6e96<\/strong>\uff1a\u63d0\u4f9b Stream3D-1M Dataset \u6a19\u8a3b\u53ca Stream3D-Bench\uff0c\u6db5\u84cb 29 \u9805\u4efb\u52d9\u3002<\/li>\n\n\n\n<li><strong>\u53ef\u8996\u5316\u8cc7\u6e90<\/strong>\uff1a\u652f\u63f4 GLB \u8207 RRD\uff0c\u65b9\u4fbf\u6aa2\u67e5\u589e\u91cf\u91cd\u5efa\u8207\u76f8\u6a5f\u8ecc\u8de1\u3002<\/li>\n\n\n\n<li><strong>\u76f8\u95dc\u6a21\u578b<\/strong>\uff1a\u5df2\u516c\u958b Stream3D-VLM-4B\uff0c\u53ef\u4f5c\u70ba\u4e86\u89e3\u6574\u9ad4\u80fd\u529b\u7684\u4e3b\u8981\u5165\u53e3\u3002<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">\u6574\u9ad4\u4f86\u770b\uff0cStream3D-VLM \u6700\u9069\u5408\u7528\u4f86\u89c0\u5bdf 3D \u591a\u6a21\u614b\u6a21\u578b\u5982\u4f55\u7531\u300c\u770b\u5b8c\u6574\u6bb5\u518d\u7b54\u300d\u8d70\u5411\u300c\u908a\u770b\u908a\u7b54\u300d\u3002\u5b83\u672a\u5fc5\u662f\u4e00\u822c\u958b\u767c\u8005\u5373\u88dd\u5373\u7528\u7684\u8f15\u91cf\u5de5\u5177\uff0c\u4f46\u5c0d\u7814\u7a76\u4e32\u6d41\u5834\u666f\u7406\u89e3\u30013D \u554f\u7b54\u3001\u7a7a\u9593\u63a8\u7406\u6d41\u7a0b\u7684\u4eba\uff0c\u9019\u500b\u9805\u76ee\u6709\u76f8\u7576\u6e05\u6670\u7684\u65b9\u5411\u8207\u5be6\u9a57\u91ce\u5fc3\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>GitHub\uff1a<\/strong> <a href=\"https:\/\/github.com\/hanxunyu\/Stream3D-VLM\" rel=\"noopener noreferrer\">https:\/\/github.com\/hanxunyu\/Stream3D-VLM<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>\u9805\u76ee\uff1a<\/strong> <a href=\"https:\/\/stream3d-vlm.github.io\/\" rel=\"noopener noreferrer\">https:\/\/stream3d-vlm.github.io\/<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u8655\u7406\u7684\u4e0d\u53ea\u662f\u5f71\u7247\uff0c\u800c\u662f\u908a\u770b\u908a\u7406\u89e3\u7a7a\u9593\u3002Stream3D-VLM\u628a\u4e32\u6d41\u756b\u9762\u3001\u5e7e\u4f55\u8cc7\u8a0a\u8207\u8a9e\u8a00\u56de\u61c9\u9023\u8d77\u4f86\u3002<\/p>\n","protected":false},"author":8,"featured_media":8938,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ai_generated_summary":"","footnotes":""},"categories":[165,172,179,119,76,149,184],"tags":[],"class_list":["post-8939","post","type-post","status-publish","format-standard","hentry","category-165","category-172","category-nvidia","category-119","category-76","category-149","category-robotic"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/8939","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/comments?post=8939"}],"version-history":[{"count":1,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/8939\/revisions"}],"predecessor-version":[{"id":8942,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/8939\/revisions\/8942"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media\/8938"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=8939"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/categories?post=8939"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/tags?post=8939"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}