{"id":1177369,"date":"2026-07-01T03:31:34","date_gmt":"2026-07-01T10:31:34","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-blog-post&p=1177369"},"modified":"2026-07-01T03:31:36","modified_gmt":"2026-07-01T10:31:36","slug":"%e5%be%ae%e8%bd%af%e4%ba%9a%e6%b4%b2%e7%a0%94%e7%a9%b6%e9%99%a2%e7%a0%94%e7%a9%b6%e6%88%90%e6%9e%9c%e8%8e%b7%e5%be%97-cvpr-2026-%e6%9c%80%e4%bd%b3%e5%ad%a6%e7%94%9f%e8%ae%ba%e6%96%87%e5%a5%96%e3%80%81","status":"publish","type":"msr-blog-post","link":"https:\/\/www.microsoft.com\/en-us\/research\/articles\/%e5%be%ae%e8%bd%af%e4%ba%9a%e6%b4%b2%e7%a0%94%e7%a9%b6%e9%99%a2%e7%a0%94%e7%a9%b6%e6%88%90%e6%9e%9c%e8%8e%b7%e5%be%97-cvpr-2026-%e6%9c%80%e4%bd%b3%e5%ad%a6%e7%94%9f%e8%ae%ba%e6%96%87%e5%a5%96%e3%80%81\/","title":{"rendered":"\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u7814\u7a76\u6210\u679c\u83b7\u5f97 CVPR 2026 \u6700\u4f73\u5b66\u751f\u8bba\u6587\u5956\u3001\u65f6\u95f4\u68c0\u9a8c\u5956\uff01"},"content":{"rendered":"\n

\u5728\u521a\u521a\u843d\u5e55\u7684\u8ba1\u7b97\u673a\u89c6\u89c9\u4e0e\u4eba\u5de5\u667a\u80fd\u9886\u57df\u6700\u5177\u5f71\u54cd\u529b\u7684\u56fd\u9645\u9876\u4f1a\u4e4b\u4e00 CVPR 2026 \u4e0a\uff0c\u6765\u81ea\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u7684\u8bba\u6587\u300aNative and Compact Structured Latents for 3D Generation\u300b\uff08\u5373TRELLIS.2\uff09<\/a>\u4ece\u5168\u7403 16,092\u7bc7\u6295\u7a3f\u4e2d\u8131\u9896\u800c\u51fa\uff0c\u8363\u81ba\u6700\u4f73\u5b66\u751f\u8bba\u6587\u5956\uff08Best Student Paper Award\uff09\u3002<\/p>\n\n\n\n

\"photo\"<\/figure>\n\n\n\n

\u4f5c\u4e3a\u5168\u65b0\u7684 3D \u8d44\u4ea7\u751f\u6210\u8303\u5f0f\uff0cTRELLIS.2\u200c\u4e0d\u4ec5\u80fd\u591f\u751f\u6210\u6db5\u76d6\u91d1\u5c5e\u3001\u5851\u6599\u3001\u73bb\u7483\u3001\u6728\u6750\u3001\u6c34\u7eb9\u7b49\u4e30\u5bcc\u6750\u8d28\u7684 3D \u7269\u4f53\uff0c\u66f4\u80fd\u5b8c\u6574\u6784\u5efa\u7269\u4f53\u5185\u90e8\u7684\u51e0\u4f55\u7ed3\u6784\u3002\u200cTRELLIS.2 \u65e2\u80fd\u5448\u73b0\u8f7b\u8584\u900f\u660e\u3001\u5c42\u6b21\u5206\u660e\u7684\u82b1\u74e3\uff0c\u4e5f\u80fd\u5728\u751f\u6210\u73bb\u7483\u74f6\u7684\u540c\u65f6\uff0c\u8ba9\u74f6\u5185\u88c5\u8f7d\u7684\u6bcf\u4e00\u4ef6\u7269\u54c1\u90fd\u6e05\u6670\u53ef\u89c1\u3002\u8fd9\u6807\u5fd7\u7740 AI \u751f\u6210\u7684 3D \u8d44\u4ea7\u8fc8\u5165\u4e86\u201c\u9ad8\u4fdd\u771f\u201d\u9636\u6bb5\u3002<\/p>\n\n\n\n

\u4e0e\u4f20\u7edf\u57fa\u4e8e\u573a\u8868\u8fbe\u7684 3D \u751f\u6210\u65b9\u6cd5\u4e0d\u540c\uff0cTRELLIS.2 \u521b\u65b0\u6027\u5730\u63d0\u51fa\u4e86\u975e\u573a\uff08field-free\uff09\u7684\u65b0\u8868\u8fbe\u2014\u2014\u7a00\u758f\u4f53\u7d20\u7ed3\u6784 O-Voxel\u3002\u8fd9\u4e00\u8868\u793a\u65b9\u6cd5\u53ef\u4ee5\u751f\u6210\u5177\u6709\u4efb\u610f\u62d3\u6251\u7ed3\u6784\u548c\u4e30\u5bcc\u6750\u8d28\u5c5e\u6027\u7684\u9ad8\u5206\u8fa8\u7387 3D \u8d44\u4ea7\uff0c\u5e76\u4e14\u5927\u5e45\u51cf\u8f7b\u5f00\u53d1\u8005\u5728\u9884\u5904\u7406\u9636\u6bb5\u7684\u8d1f\u62c5\u3002\u4e0d\u4ec5\u5982\u6b64\uff0cTRELLIS.2 \u8fd8\u5b9e\u73b0\u4e8616\u500d\u7684\u7a7a\u95f4\u538b\u7f29\uff0c\u8ba9\u62e5\u670940\u4ebf\u53c2\u6570\u7684\u5927\u578b\u751f\u6210\u6a21\u578b\u4e5f\u80fd\u9ad8\u6548\u5b8c\u6210\u8bad\u7ec3\u548c\u63a8\u7406\u3002\u5728\u5b9e\u9645\u6027\u80fd\u8868\u73b0\u4e0a\uff0c\u5176\u751f\u6210512\u00b3\u5206\u8fa8\u7387\u7684\u5168\u7eb9\u7406\u8d44\u4ea7\u4ec5\u9700\u7ea63\u79d2\u3002<\/p>\n\n\n\n

\u6b64\u5916\uff0c\u672c\u5c4a\u5927\u4f1a\u8fd8\u5c06\u81f3\u9ad8\u8363\u8a89\u4e4b\u4e00 Longuet-Higgins Prize \uff08\u65f6\u95f4\u68c0\u9a8c\u5956\uff09\u9881\u53d1\u7ed9\u4e86 2015 \u5e74\u7531\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u56e2\u961f\u53d1\u8868\u7684\u7ecf\u5178\u8bba\u6587\u300aDeep Residual Learning for Image Recognition\u300b\uff08\u5373 ResNet\uff09\u3002\u8be5\u5de5\u4f5c\u4f7f\u795e\u7ecf\u7f51\u7edc\u8fbe\u5230\u4e86\u524d\u6240\u672a\u6709\u7684\u6df1\u5ea6\uff0c\u83b7\u5f97\u4e86\u6b64\u524d\u96be\u4ee5\u5b9e\u73b0\u7684\u80fd\u529b\uff0c\u8fdb\u800c\u4fc3\u6210\u4e86\u5305\u62ecAlphaGo\uff0cAlphaFold \u548c ChatGPT\u5728\u5185\u7684\u591a\u4e2a\u7a81\u7834\u6027\u6210\u679c\u3002<\/p>\n\n\n\n

\"photo\"<\/figure>\n\n\n\n

\u4ece ResNet\u7684\u5f00\u521b\u6027\u8d21\u732e\uff0c\u5230 TRELLIS.2 \u5bf9\u4e09\u7ef4\u751f\u6210\u4e0e\u7a7a\u95f4\u667a\u80fd\u8fb9\u754c\u7684\u6301\u7eed\u62d3\u5c55\uff0c\u8fd9\u4e9b\u7814\u7a76\u5de5\u4f5c\u4e0d\u4ec5\u5f97\u5230\u4e86\u56fd\u9645\u5b66\u672f\u754c\u6700\u9ad8\u89c4\u683c\u7684\u8ba4\u53ef\uff0c\u66f4\u5145\u5206\u4f53\u73b0\u4e86\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u957f\u671f\u575a\u6301\u57fa\u7840\u7814\u7a76\u3001\u6301\u7eed\u63a8\u52a8\u521b\u65b0\u7684\u79d1\u7814\u7cbe\u795e\u3002\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u5c06\u7ee7\u7eed\u63a2\u7d22\u80fd\u591f\u5b9a\u4e49\u672a\u6765\u6280\u672f\u8303\u5f0f\u7684\u5173\u952e\u95ee\u9898\uff0c\u4e3a\u5168\u7403 AI \u53d1\u5c55\u8d21\u732e\u66f4\u591a\u5177\u6709\u957f\u671f\u5f71\u54cd\u529b\u7684\u521b\u65b0\u6210\u679c\u3002<\/p>\n\n\n\n

\u6700\u540e\uff0c\u5411\u672c\u5c4aCVPR\u5927\u4f1a\u7684\u6240\u6709\u83b7\u5956\u8005\u81f4\u4ee5\u8bda\u631a\u795d\u8d3a\uff01

\u76f8\u5173\u8bba\u6587\uff1a
Native and Compact Structured Latents for 3D Generation<\/p>\n\n\n\n

\u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/openaccess.thecvf.com\/content\/CVPR2026\/papers\/Xiang_Native_and_Compact_Structured_Latents_for_3D_Generation_CVPR_2026_paper.pdf<\/p>\n\n\n\n

Deep Residual Learning for Image Recognition<\/p>\n\n\n\n

\u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/1512.03385<\/p>\n","protected":false},"excerpt":{"rendered":"

\u5728\u521a\u521a\u843d\u5e55\u7684\u8ba1\u7b97\u673a\u89c6\u89c9\u4e0e\u4eba\u5de5\u667a\u80fd\u9886\u57df\u6700\u5177\u5f71\u54cd\u529b\u7684\u56fd\u9645\u9876\u4f1a\u4e4b\u4e00 CVPR 2026 \u4e0a\uff0c\u6765\u81ea\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u7684\u8bba\u6587\u300aNative and Compact Structured Latents for 3D Generation\u300b\uff08\u5373TRELLIS.2\uff09\u4ece\u5168\u7403 16,092\u7bc7\u6295\u7a3f\u4e2d\u8131\u9896\u800c\u51fa\uff0c\u8363\u81ba\u6700\u4f73\u5b66\u751f\u8bba\u6587\u5956\uff08Best Student Paper Award\uff09\u3002 \u4f5c\u4e3a\u5168\u65b0\u7684 3D \u8d44\u4ea7\u751f\u6210\u8303\u5f0f\uff0cTRELLIS.2\u200c\u4e0d\u4ec5\u80fd\u591f\u751f\u6210\u6db5\u76d6\u91d1\u5c5e\u3001\u5851\u6599\u3001\u73bb\u7483\u3001\u6728\u6750\u3001\u6c34\u7eb9\u7b49\u4e30\u5bcc\u6750\u8d28\u7684 3D \u7269\u4f53\uff0c\u66f4\u80fd\u5b8c\u6574\u6784\u5efa\u7269\u4f53\u5185\u90e8\u7684\u51e0\u4f55\u7ed3\u6784\u3002\u200cTRELLIS.2 \u65e2\u80fd\u5448\u73b0\u8f7b\u8584\u900f\u660e\u3001\u5c42\u6b21\u5206\u660e\u7684\u82b1\u74e3\uff0c\u4e5f\u80fd\u5728\u751f\u6210\u73bb\u7483\u74f6\u7684\u540c\u65f6\uff0c\u8ba9\u74f6\u5185\u88c5\u8f7d\u7684\u6bcf\u4e00\u4ef6\u7269\u54c1\u90fd\u6e05\u6670\u53ef\u89c1\u3002\u8fd9\u6807\u5fd7\u7740 AI \u751f\u6210\u7684 3D \u8d44\u4ea7\u8fc8\u5165\u4e86\u201c\u9ad8\u4fdd\u771f\u201d\u9636\u6bb5\u3002 \u4e0e\u4f20\u7edf\u57fa\u4e8e\u573a\u8868\u8fbe\u7684 3D \u751f\u6210\u65b9\u6cd5\u4e0d\u540c\uff0cTRELLIS.2 \u521b\u65b0\u6027\u5730\u63d0\u51fa\u4e86\u975e\u573a\uff08field-free\uff09\u7684\u65b0\u8868\u8fbe\u2014\u2014\u7a00\u758f\u4f53\u7d20\u7ed3\u6784 O-Voxel\u3002\u8fd9\u4e00\u8868\u793a\u65b9\u6cd5\u53ef\u4ee5\u751f\u6210\u5177\u6709\u4efb\u610f\u62d3\u6251\u7ed3\u6784\u548c\u4e30\u5bcc\u6750\u8d28\u5c5e\u6027\u7684\u9ad8\u5206\u8fa8\u7387 3D \u8d44\u4ea7\uff0c\u5e76\u4e14\u5927\u5e45\u51cf\u8f7b\u5f00\u53d1\u8005\u5728\u9884\u5904\u7406\u9636\u6bb5\u7684\u8d1f\u62c5\u3002\u4e0d\u4ec5\u5982\u6b64\uff0cTRELLIS.2 \u8fd8\u5b9e\u73b0\u4e8616\u500d\u7684\u7a7a\u95f4\u538b\u7f29\uff0c\u8ba9\u62e5\u670940\u4ebf\u53c2\u6570\u7684\u5927\u578b\u751f\u6210\u6a21\u578b\u4e5f\u80fd\u9ad8\u6548\u5b8c\u6210\u8bad\u7ec3\u548c\u63a8\u7406\u3002\u5728\u5b9e\u9645\u6027\u80fd\u8868\u73b0\u4e0a\uff0c\u5176\u751f\u6210512\u00b3\u5206\u8fa8\u7387\u7684\u5168\u7eb9\u7406\u8d44\u4ea7\u4ec5\u9700\u7ea63\u79d2\u3002 \u6b64\u5916\uff0c\u672c\u5c4a\u5927\u4f1a\u8fd8\u5c06\u81f3\u9ad8\u8363\u8a89\u4e4b\u4e00 Longuet-Higgins Prize \uff08\u65f6\u95f4\u68c0\u9a8c\u5956\uff09\u9881\u53d1\u7ed9\u4e86 2015 \u5e74\u7531\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u56e2\u961f\u53d1\u8868\u7684\u7ecf\u5178\u8bba\u6587\u300aDeep Residual Learning for Image Recognition\u300b\uff08\u5373 ResNet\uff09\u3002\u8be5\u5de5\u4f5c\u4f7f\u795e\u7ecf\u7f51\u7edc\u8fbe\u5230\u4e86\u524d\u6240\u672a\u6709\u7684\u6df1\u5ea6\uff0c\u83b7\u5f97\u4e86\u6b64\u524d\u96be\u4ee5\u5b9e\u73b0\u7684\u80fd\u529b\uff0c\u8fdb\u800c\u4fc3\u6210\u4e86\u5305\u62ecAlphaGo\uff0cAlphaFold \u548c ChatGPT\u5728\u5185\u7684\u591a\u4e2a\u7a81\u7834\u6027\u6210\u679c\u3002 \u4ece ResNet\u7684\u5f00\u521b\u6027\u8d21\u732e\uff0c\u5230 TRELLIS.2 \u5bf9\u4e09\u7ef4\u751f\u6210\u4e0e\u7a7a\u95f4\u667a\u80fd\u8fb9\u754c\u7684\u6301\u7eed\u62d3\u5c55\uff0c\u8fd9\u4e9b\u7814\u7a76\u5de5\u4f5c\u4e0d\u4ec5\u5f97\u5230\u4e86\u56fd\u9645\u5b66\u672f\u754c\u6700\u9ad8\u89c4\u683c\u7684\u8ba4\u53ef\uff0c\u66f4\u5145\u5206\u4f53\u73b0\u4e86\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u957f\u671f\u575a\u6301\u57fa\u7840\u7814\u7a76\u3001\u6301\u7eed\u63a8\u52a8\u521b\u65b0\u7684\u79d1\u7814\u7cbe\u795e\u3002\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\u5c06\u7ee7\u7eed\u63a2\u7d22\u80fd\u591f\u5b9a\u4e49\u672a\u6765\u6280\u672f\u8303\u5f0f\u7684\u5173\u952e\u95ee\u9898\uff0c\u4e3a\u5168\u7403 AI \u53d1\u5c55\u8d21\u732e\u66f4\u591a\u5177\u6709\u957f\u671f\u5f71\u54cd\u529b\u7684\u521b\u65b0\u6210\u679c\u3002 \u6700\u540e\uff0c\u5411\u672c\u5c4aCVPR\u5927\u4f1a\u7684\u6240\u6709\u83b7\u5956\u8005\u81f4\u4ee5\u8bda\u631a\u795d\u8d3a\uff01 \u76f8\u5173\u8bba\u6587\uff1aNative […]<\/p>\n","protected":false},"author":44093,"featured_media":1177373,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-content-parent":1012650,"msr_hide_image_in_river":0,"footnotes":""},"research-area":[13556],"msr-locale":[268881],"msr-post-option":[],"class_list":["post-1177369","msr-blog-post","type-msr-blog-post","status-publish","has-post-thumbnail","hentry","msr-research-area-artificial-intelligence","msr-locale-zh_cn"],"msr_assoc_parent":{"id":1012650,"type":"lab"},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-blog-post\/1177369","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-blog-post"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-blog-post"}],"author":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/users\/44093"}],"version-history":[{"count":2,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-blog-post\/1177369\/revisions"}],"predecessor-version":[{"id":1177378,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-blog-post\/1177369\/revisions\/1177378"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/1177373"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1177369"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1177369"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1177369"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1177369"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}