
{"id":10203,"date":"2026-07-14T16:09:44","date_gmt":"2026-07-14T08:09:44","guid":{"rendered":"https:\/\/infernews.com\/blog\/controllable-virtual-try-on-via-visual-instance-prompt-segmentation\/"},"modified":"2026-07-14T16:09:44","modified_gmt":"2026-07-14T08:09:44","slug":"controllable-virtual-try-on-via-visual-instance-prompt-segmentation","status":"publish","type":"post","link":"https:\/\/infernews.com\/blog\/controllable-virtual-try-on-via-visual-instance-prompt-segmentation\/","title":{"rendered":"CtrlVTON \u628a\u865b\u64ec\u8a66\u7a7f\u8b8a\u6210\u53ef\u63a7\u7de8\u8f2f"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/07\/teaser-7.jpg\" alt=\"NXN LABS\"><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">\u7db2\u8cfc\u8a66\u886b\u6700\u5e38\u898b\u7684\u843d\u5dee\uff0c\u4e0d\u662f\u50cf\u4e0d\u50cf\uff0c\u800c\u662f\u4f60\u6839\u672c\u7121\u6cd5\u6307\u5b9a\u4ef6\u886b\u8981\u9ede\u6a23\u8457\u3002CtrlVTON \u5c6c\u65bc\u5f71\u50cf\u751f\u6210\u8207\u865b\u64ec\u8a66\u7a7f\u9805\u76ee\uff0c\u8655\u7406\u7684\u662f Virtual try-on (VTO) \u9577\u671f\u6b20\u7f3a\u7684\u53ef\u63a7\u6027\uff1a\u540c\u4e00\u4ef6\u886b\u53ef\u4ee5\u9b06\u8eab\u6216\u8cbc\u8eab\u3001\u675f\u5165\u6216\u653e\u51fa\u3001\u6253\u958b\u6216\u6263\u4e0a\uff0c\u751a\u81f3\u6539\u8b8a\u7a7f\u8457\u4f4d\u7f6e\u8207\u758a\u7a7f\u65b9\u5f0f\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u73fe\u6709\u505a\u6cd5\u591a\u6578\u628a VTO \u7576\u6210 inpainting \u554f\u984c\uff0c\u64c5\u9577\u88dc\u756b\u8863\u7269\uff0c\u537b\u4e0d\u5bb9\u6613\u7cbe\u6e96\u8ddf\u4f4f\u4f7f\u7528\u8005\u6307\u5b9a\u7684\u7248\u578b\u540c\u4f4d\u7f6e\u3002CtrlVTON \u6539\u7528 image editing framing\uff0c\u518d\u914d\u5408 segmentation mask \u505a pixel-level control\uff0c\u91cd\u9ede\u4e0d\u662f\u55ae\u7d14\u751f\u6210\u5f97\u81ea\u7136\uff0c\u800c\u662f\u4ee4\u7248\u9762\u5b89\u6392\u771f\u6b63\u807d\u8a71\u3002\u8ad6\u6587\u4ea6\u660e\u78ba\u6307\u51fa\uff0c\u5b83\u5728\u670d\u88dd\u5fe0\u5be6\u5ea6\u63a5\u8fd1\u5f37\u52e2 proprietary editing systems \u7684\u540c\u6642\uff0c\u5c0d\u4f7f\u7528\u8005\u63d0\u4f9b\u7684\u5e03\u5c40\u8ddf\u96a8\u5f97\u66f4\u6e96\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u652f\u6490\u9019\u5957\u65b9\u6cd5\u7684\u53e6\u4e00\u534a\uff0c\u662f Visual-Instance-Prompt Segmentation \u8207\u5c0d\u61c9\u6a21\u578b VIP-SAM\u3002\u4ee5\u5f80\u5e38\u898b\u7684\u662f category-level visual-reference segmentation\uff0c\u4f46\u4f5c\u8005\u91dd\u5c0d\u7684\u662f instance-level \u4efb\u52d9\uff1a\u5148\u7d66\u4f60\u4e00\u5f35\u5e73\u62cd\u670d\u88dd\u5716\uff0c\u518d\u5f9e\u4eba\u7269\u7a7f\u8457\u7167\u7247\u4e2d\u5206\u51fa\u540c\u4e00\u4ef6\u8863\u7269\u3002\u9019\u4e00\u6b65\u5c0d\u540c\u985e\u8863\u7269\u5e72\u64fe\u3001\u906e\u64cb\uff0c\u4ee5\u53ca\u5e03\u6599\u8b8a\u5f62\u90fd\u66f4\u654f\u611f\uff0c\u6240\u4ee5 VIP-SAM \u6703\u628a support features \u63d0\u65e9\u6ce8\u5165 SAM backbone\uff0c\u76ee\u6a19\u662f\u4ee4\u5b9a\u4f4d\u66f4\u7a69\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u628a VTO \u7531 inpainting \u8f49\u6210 image editing\uff0c\u63a7\u5236\u7c92\u5ea6\u66f4\u7d30<\/li>\n<li>\u7528 segmentation masks \u6307\u5b9a\u670d\u88dd\u5927\u5c0f\u3001\u98a8\u683c\u8207\u8eab\u9ad4\u4e0a\u7684\u4f4d\u7f6e<\/li>\n<li>VIP-SAM \u5148\u89e3\u6c7a\u300c\u76f8\u4e2d\u7a76\u7adf\u4fc2\u908a\u4e00\u4ef6\u886b\u300d\u7684 instance-level \u5206\u5272\u554f\u984c<\/li>\n<li>\u4e00\u500b\u6846\u67b6\u5167\u8655\u7406 garment swapping\u3001layering\u3001selective switching\u3001multi-garment composition<\/li>\n<li>\u5df2\u516c\u958b VITON-HD-edit dataset\uff0c\u4e26\u6709 arXiv \u8ad6\u6587\u53ef\u4ea4\u53c9\u6838\u5c0d\u65b9\u6cd5\u8207\u7d50\u679c<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">\u7406\u89e3\u9019\u500b\u9805\u76ee\u7684\u8f03\u5408\u7406\u65b9\u5f0f\uff0c\u4e0d\u662f\u628a\u5b83\u7576\u6210\u5373\u88dd\u5373\u7528\u6210\u54c1\uff0c\u800c\u662f\u7814\u7a76\u578b\u539f\u578b\u52a0\u8cc7\u6599\u96c6\u7d44\u5408\u3002\uff0c\u6e2c\u8a66\u8207\u90e8\u7f72\u6703\u570d\u7e5e\u8ad6\u6587\u3001GitHub \u5132\u5b58\u5eab\uff0c\u4ee5\u53ca Hugging Face \u4e0a\u7684 VITON-HD-edit dataset \u5c55\u958b\uff1b\u8f03\u9069\u5408\u505a\u53ef\u63a7\u8a66\u7a7f\u7814\u7a76\u3001\u96fb\u5546\u5f71\u50cf\u6d41\u7a0b\u9a57\u8b49\uff0c\u6216\u60f3\u6bd4\u8f03 LoRA \u5f0f\u8f15\u91cf\u5fae\u8abf\u80fd\u5426\u53d6\u4ee3\u5927\u578b\u5c01\u9589\u7de8\u8f2f\u670d\u52d9\u7684\u5718\u968a\u3002\u9650\u5236\u4e5f\u5f88\u6e05\u695a\uff1aProject Page \u5c1a\u672a\u516c\u958b\uff0cREADME \u91cb\u51fa\u7684\u5b89\u88dd\u8207\u63a8\u7406\u7d30\u7bc0\u4ecd\u4e0d\u5b8c\u6574\uff0c\u73fe\u968e\u6bb5\u66f4\u9069\u5408\u6709\u5f71\u50cf\u6a21\u578b\u57fa\u790e\u7684\u4eba\u5148\u8ddf\u8ad6\u6587\u8a2d\u5b9a\u91cd\u73fe\uff0c\u518d\u8a55\u4f30\u80fd\u5426\u63a5\u5165\u7522\u54c1\u6d41\u7a0b\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/github.com\/nxnai\/CtrlVTON\" rel=\"noopener noreferrer\" target=\"_blank\"><strong>GitHub<\/strong><\/a> \u00b7 <a href=\"https:\/\/arxiv.org\/pdf\/2607.09362\" rel=\"noopener noreferrer\" target=\"_blank\"><strong>Paper<\/strong><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u60f3\u50cf\u7db2\u8cfc\u8a66\u886b\u5514\u53ea\u4fc2\u63db\u4ef6\u886b\u4e0a\u8eab\uff0c\u9023\u9b06\u8eab\u3001\u675f\u5165\u8932\u982d\u540c\u758a\u7a7f\u4f4d\u7f6e\u90fd\u53ef\u4ee5\u6307\u5b9a\u3002CtrlVTON\u7784\u6e96\u7684\uff0c\u6b63\u6b63\u662f\u9019\u7a2e\u4ee5\u5f80 VTO \u5f88\u96e3\u8655\u7406\u7684\u63a7\u5236\u611f\u3002<\/p>\n","protected":false},"author":8,"featured_media":10202,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ai_generated_summary":"","footnotes":""},"categories":[133,30,157,120,199],"tags":[],"class_list":["post-10203","post","type-post","status-publish","format-standard","hentry","category-133","category-image","category-157","category-120","category-dataset-"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/10203","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=10203"}],"version-history":[{"count":0,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/10203\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media\/10202"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=10203"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/categories?post=10203"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/tags?post=10203"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}