DEtection with TRansformers 阅读笔记<\/title>\n <style>\n\/* From extension vscode.github *\/\n\/*---------------------------------------------------------------------------------------------\n * Copyright (c) Microsoft Corporation. All rights reserved.\n * Licensed under the MIT License. See License.txt in the project root for license information.\n *--------------------------------------------------------------------------------------------*\/\n\n.vscode-dark img[src$=\\#gh-light-mode-only],\n.vscode-light img[src$=\\#gh-dark-mode-only] {\n\tdisplay: none;\n}\n\n<\/style>\n \n <link rel=\"stylesheet\" href=\"https:\/\/cdn.jsdelivr.net\/gh\/Microsoft\/vscode\/extensions\/markdown-language-features\/media\/markdown.css\">\n<link rel=\"stylesheet\" href=\"https:\/\/cdn.jsdelivr.net\/gh\/Microsoft\/vscode\/extensions\/markdown-language-features\/media\/highlight.css\">\n<style>\n body {\n font-family: -apple-system, BlinkMacSystemFont, 'Segoe WPC', 'Segoe UI', system-ui, 'Ubuntu', 'Droid Sans', sans-serif;\n font-size: 14px;\n line-height: 1.6;\n }\n <\/style>\n <style>\n.task-list-item {\n list-style-type: none;\n}\n\n.task-list-item-checkbox {\n margin-left: -20px;\n vertical-align: middle;\n pointer-events: none;\n}\n<\/style>\n \n <\/head>\n <body class=\"vscode-body vscode-light\">\n <h1 id=\"detection-with-transformers-\u9605\u8bfb\u7b14\u8bb0\">DEtection with TRansformers \u9605\u8bfb\u7b14\u8bb0<\/h1>\n<h2 id=\"\u6a21\u578b\u7ed3\u6784\">\u6a21\u578b\u7ed3\u6784<\/h2>\n<h3 id=\"cnn-backbone\">CNN Backbone<\/h3>\n<p>\u5e38\u89c4\u505a\u6cd5\uff0c \u7528\u6765\u7279\u5f81\u63d0\u53d6\u3002\u8f93\u51fa\u4e00\u4e2a C * HW \u7684\u7279\u5f81\u56fe\u3002<\/p>\n<h3 id=\"transformer\">Transformer<\/h3>\n<ul>\n<li>\u591a\u5934\u6ce8\u610f\u529b\u673a\u5236<\/li>\n<li>\u628a <code>pos_embed<\/code> \u7ed9 sum \u5230 Encoder \u7684 query \u548c key \u91cc<\/li>\n<li>Decoder \u91cc\u56fa\u5b9a\u67e5 N = 100 \u4e2a token<\/li>\n<\/ul>\n<h3 id=\"ffn\">FFN<\/h3>\n<p>\u5b9e\u9645\u4e0a\u5c31\u662f 1*1 \u7684\u5377\u79ef\u3002<\/p>\n<h2 id=\"\u6570\u636e\">\u6570\u636e<\/h2>\n<p>COCO Dataset<\/p>\n<h2 id=\"\u8bad\u7ec3\">\u8bad\u7ec3<\/h2>\n<h3 id=\"\u8ba1\u7b97\u635f\u5931\">\u8ba1\u7b97\u635f\u5931<\/h3>\n<p>\u7531\u4e8e\u4ea7\u51fa 100 \u4e2a\u68c0\u6d4b\u7ed3\u679c\uff0c\u6545\u800c\u548c\u4e8b\u5b9e\u505a\u4e8c\u5206\u56fe\u5339\u914d\uff0c\u8dd1\u4e2a Hungarian \u7b97\u6cd5\u5373\u53ef\u83b7\u5f97\u5339\u914d\u5173\u7cfb\u3002<\/p>\n<h1 id=\"dino-\u9605\u8bfb\u7b14\u8bb0\">DINO \u9605\u8bfb\u7b14\u8bb0<\/h1>\n<h2 id=\"preliminaries\">Preliminaries<\/h2>\n<p>\u8bba\u6587\u63d0\u5230\u4e24\u4e2a\u5bf9\u5b83\u5f71\u54cd\u6bd4\u8f83\u5927\u7684\u5de5\u4f5c\uff1a<\/p>\n<ul>\n<li>DAB-DETR \u5c06 positional query \u63cf\u8ff0\u6210 R^4 \u91cc\u7684\u4e00\u4e2a\u5411\u91cf (x, y, w, h)<\/li>\n<li>DN-DETR \u8ba4\u4e3a\u6536\u655b\u7f13\u6162\u53ef\u80fd\u662f\u56e0\u4e3a\u4e8c\u5206\u56fe\u5339\u914d\u4e0d\u591f\u7a33\u5b9a\uff0c\u56e0\u6b64\u5f15\u5165 denosing \uff0c\u5c06\u5e26\u566a\u58f0\u7684 label \u548c box \u9001\u5165 DETR<\/li>\n<li>Deformable DETR \u8ba9 attention \u5173\u6ce8 sampling points \u5468\u56f4\u7684\u4e00\u5c0f\u90e8\u4efd\u70b9\u3002\u9996\u5148\u5b83\u5f15\u5165\u4e86 query selection \u6280\u672f\uff08\u76f4\u63a5\u4ece encoder \u7684\u8f93\u51fa\u63d0\u53d6 features \u548c reference boxes\uff09\uff1b\u5176\u6b21\u662f\u5b83\u7528\u4e86\u6240\u8c13\u68af\u5ea6\u5206\u79bb\u6280\u672f\uff0c\u8fd9\u91cc\u53eb\u5b83 look forward once<\/li>\n<\/ul>\n<h2 id=\"backbone\">Backbone<\/h2>\n<p>\u5206\u522b\u7528 ResNet-50 \u548c SwinL \u4f5c\u4e3a backbone<\/p>\n<h2 id=\"\u6a21\u578b\">\u6a21\u578b<\/h2>\n<ul>\n<li>look forward twice: \u4e00\u4e2a\u7c7b\u4f3c residual connection \u7684\u64cd\u4f5c\uff0c\u7528\u6765\u51cf\u5c11\u68af\u5ea6\u6d88\u5931\u7684\u98ce\u9669<\/li>\n<li>Mixed Query Selection: \u9009\u62e9 top-k \u4e2a feature \u4f20\u5165 decoder<\/li>\n<\/ul>\n<h2 id=\"\u8bad\u7ec3-1\">\u8bad\u7ec3<\/h2>\n<ul>\n<li>Contrastive DeNoising Training (CDN) \uff0c\u5c06 Ground Truth \u5185\u7684\u4f5c\u4e3a\u6b63\u6837\u672c\uff0c\u5916\u7684\u4f5c\u4e3a\u8d1f\u6837\u672c\uff0c\u53ef\u4ee5\u907f\u514d\u91cd\u590d\u9884\u6d4b\u3002<\/li>\n<\/ul>\n<h1 id=\"grounding-dino-\u9605\u8bfb\u7b14\u8bb0\">Grounding DINO \u9605\u8bfb\u7b14\u8bb0<\/h1>\n<h2 id=\"preliminaries-1\">Preliminaries<\/h2>\n<p>\u63d0\u51fa\u4e00\u4e2a\u8303\u5f0f\uff1aBackbone \u7528\u4e8e feature extraction, neck \u7528\u4e8e feature enchancement, head \u7528\u4e8e region refinement(box prediction)<\/p>\n<h2 id=\"\u6a21\u578b-1\">\u6a21\u578b<\/h2>\n<p>image \u7ecf\u8fc7\u4e00\u4e2a Swin-T \u4f5c\u4e3a backbone, \u800c text \u5219\u7ecf\u8fc7\u4e00\u4e2a BERT<\/p>\n<p>\u7136\u540e\u662f\u4e00\u4e2a Feature Enhance Layer, SAM \u548c\u5b83\u6709\u70b9\u76f8\u4f3c\uff0c\u90fd\u662f image \u548c text (\u5728 SAM \u91cc\u662f prompt) \u5404\u7ecf\u8fc7\u4e00\u4e2a self-attention \/ deformable self-attention \u4ee5\u540e\u505a\u4e00\u4e2a\u53cc\u5411\u7684 cross-attention, \u7136\u540e\u5404 FFN \u4ee5\u540e\u8f93\u51fa\u3002<\/p>\n<p>\u63a5\u7740\u662f Language-Guided Query Selection, \u8fd9\u4e2a\u90e8\u5206\u9996\u5148\u5c06 image \u548c text \u7684 feature \u6cbf\u7b2c\u4e09\u7ef4\u5377\u79ef\uff0c\u53d6 max \u4f5c\u4e3a img feat\uff0c\u6700\u540e\u53d6 topk \u8fd4\u56de\u3002<\/p>\n<p>\u63a5\u7740\u662f Cross-Modality Decoder, \u5c06 Query \u7ecf\u8fc7 self-attention \u4ee5\u540e\uff0c\u53d6 Q \u5206\u522b\u548c Image Features, Text Features \u7684 K, V \u505a Cross-Attention \uff0c\u518d\u8fc7\u4e00\u4e2a FFN \u4ee5\u540e\u8f93\u51fa\u3002<\/p>\n<h2 id=\"\u8f93\u5165\">\u8f93\u5165<\/h2>\n<p>\u5904\u7406 Text Prompt: \u4ee5\u5f80\u91c7\u7528\u4e24\u79cd text prompt, \u4f5c\u8005\u79f0\u4e4b\u4e3a Sentence Level \u548c Word Level; \u5728\u8fd9\u91cc\u5219\u4f7f\u7528\u4e86 Sub-sentence Level<\/p>\n<h2 id=\"\u8bad\u7ec3-2\">\u8bad\u7ec3<\/h2>\n<ul>\n<li>\u7c7b\u4f3c DETR, \u4ecd\u7136\u662f\u5148\u4e8c\u5206\u56fe\u5339\u914d\u518d\u8ba1\u7b97\u635f\u5931<\/li>\n<\/ul>\n<h2 id=\"ablations\">Ablations<\/h2>\n<p>\u6d88\u878d\u5b9e\u9a8c\u8ba4\u4e3a Encoder fusion \u662f\u6700\u91cd\u8981\u7684\u90e8\u4efd<\/p>\n\n \n \n <\/body>\n <\/html>\n","protected":false},"excerpt":{"rendered":"<p>DEtection with TRansformers 阅读笔&#x … <\/p>\n<p class=\"link-more\"><a href=\"http:\/\/SmokeyDays.top\/wordpress\/2023\/07\/15\/grounding-dino-%e9%98%85%e8%af%bb%e7%ac%94%e8%ae%b0\/\" class=\"more-link\">\u7ee7\u7eed\u9605\u8bfb<span class=\"screen-reader-text\">\u201c(grounding) DINO \u9605\u8bfb\u7b14\u8bb0\u201d<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[131,130,135],"tags":[],"_links":{"self":[{"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/posts\/1074"}],"collection":[{"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/comments?post=1074"}],"version-history":[{"count":1,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/posts\/1074\/revisions"}],"predecessor-version":[{"id":1075,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/posts\/1074\/revisions\/1075"}],"wp:attachment":[{"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/media?parent=1074"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/categories?post=1074"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/tags?post=1074"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}

{"id":1074,"date":"2023-07-15T14:46:06","date_gmt":"2023-07-15T06:46:06","guid":{"rendered":"http:\/\/SmokeyDays.top\/wordpress\/?p=1074"},"modified":"2023-07-15T14:46:06","modified_gmt":"2023-07-15T06:46:06","slug":"grounding-dino-%e9%98%85%e8%af%bb%e7%ac%94%e8%ae%b0","status":"publish","type":"post","link":"http:\/\/SmokeyDays.top\/wordpress\/2023\/07\/15\/grounding-dino-%e9%98%85%e8%af%bb%e7%ac%94%e8%ae%b0\/","title":{"rendered":"(grounding) DINO \u9605\u8bfb\u7b14\u8bb0"},"content":{"rendered":"\n\n \n \n \n DEtection with TRansformers 阅读笔记<\/title>\n <style>\n\/* From extension vscode.github *\/\n\/*---------------------------------------------------------------------------------------------\n * Copyright (c) Microsoft Corporation. All rights reserved.\n * Licensed under the MIT License. See License.txt in the project root for license information.\n *--------------------------------------------------------------------------------------------*\/\n\n.vscode-dark img[src$=\\#gh-light-mode-only],\n.vscode-light img[src$=\\#gh-dark-mode-only] {\n\tdisplay: none;\n}\n\n<\/style>\n \n <link rel=\"stylesheet\" href=\"https:\/\/cdn.jsdelivr.net\/gh\/Microsoft\/vscode\/extensions\/markdown-language-features\/media\/markdown.css\">\n<link rel=\"stylesheet\" href=\"https:\/\/cdn.jsdelivr.net\/gh\/Microsoft\/vscode\/extensions\/markdown-language-features\/media\/highlight.css\">\n<style>\n body {\n font-family: -apple-system, BlinkMacSystemFont, 'Segoe WPC', 'Segoe UI', system-ui, 'Ubuntu', 'Droid Sans', sans-serif;\n font-size: 14px;\n line-height: 1.6;\n }\n <\/style>\n <style>\n.task-list-item {\n list-style-type: none;\n}\n\n.task-list-item-checkbox {\n margin-left: -20px;\n vertical-align: middle;\n pointer-events: none;\n}\n<\/style>\n \n <\/head>\n <body class=\"vscode-body vscode-light\">\n <h1 id=\"detection-with-transformers-\u9605\u8bfb\u7b14\u8bb0\">DEtection with TRansformers \u9605\u8bfb\u7b14\u8bb0<\/h1>\n<h2 id=\"\u6a21\u578b\u7ed3\u6784\">\u6a21\u578b\u7ed3\u6784<\/h2>\n<h3 id=\"cnn-backbone\">CNN Backbone<\/h3>\n<p>\u5e38\u89c4\u505a\u6cd5\uff0c \u7528\u6765\u7279\u5f81\u63d0\u53d6\u3002\u8f93\u51fa\u4e00\u4e2a C * HW \u7684\u7279\u5f81\u56fe\u3002<\/p>\n<h3 id=\"transformer\">Transformer<\/h3>\n<ul>\n<li>\u591a\u5934\u6ce8\u610f\u529b\u673a\u5236<\/li>\n<li>\u628a <code>pos_embed<\/code> \u7ed9 sum \u5230 Encoder \u7684 query \u548c key \u91cc<\/li>\n<li>Decoder \u91cc\u56fa\u5b9a\u67e5 N = 100 \u4e2a token<\/li>\n<\/ul>\n<h3 id=\"ffn\">FFN<\/h3>\n<p>\u5b9e\u9645\u4e0a\u5c31\u662f 1*1 \u7684\u5377\u79ef\u3002<\/p>\n<h2 id=\"\u6570\u636e\">\u6570\u636e<\/h2>\n<p>COCO Dataset<\/p>\n<h2 id=\"\u8bad\u7ec3\">\u8bad\u7ec3<\/h2>\n<h3 id=\"\u8ba1\u7b97\u635f\u5931\">\u8ba1\u7b97\u635f\u5931<\/h3>\n<p>\u7531\u4e8e\u4ea7\u51fa 100 \u4e2a\u68c0\u6d4b\u7ed3\u679c\uff0c\u6545\u800c\u548c\u4e8b\u5b9e\u505a\u4e8c\u5206\u56fe\u5339\u914d\uff0c\u8dd1\u4e2a Hungarian \u7b97\u6cd5\u5373\u53ef\u83b7\u5f97\u5339\u914d\u5173\u7cfb\u3002<\/p>\n<h1 id=\"dino-\u9605\u8bfb\u7b14\u8bb0\">DINO \u9605\u8bfb\u7b14\u8bb0<\/h1>\n<h2 id=\"preliminaries\">Preliminaries<\/h2>\n<p>\u8bba\u6587\u63d0\u5230\u4e24\u4e2a\u5bf9\u5b83\u5f71\u54cd\u6bd4\u8f83\u5927\u7684\u5de5\u4f5c\uff1a<\/p>\n<ul>\n<li>DAB-DETR \u5c06 positional query \u63cf\u8ff0\u6210 R^4 \u91cc\u7684\u4e00\u4e2a\u5411\u91cf (x, y, w, h)<\/li>\n<li>DN-DETR \u8ba4\u4e3a\u6536\u655b\u7f13\u6162\u53ef\u80fd\u662f\u56e0\u4e3a\u4e8c\u5206\u56fe\u5339\u914d\u4e0d\u591f\u7a33\u5b9a\uff0c\u56e0\u6b64\u5f15\u5165 denosing \uff0c\u5c06\u5e26\u566a\u58f0\u7684 label \u548c box \u9001\u5165 DETR<\/li>\n<li>Deformable DETR \u8ba9 attention \u5173\u6ce8 sampling points \u5468\u56f4\u7684\u4e00\u5c0f\u90e8\u4efd\u70b9\u3002\u9996\u5148\u5b83\u5f15\u5165\u4e86 query selection \u6280\u672f\uff08\u76f4\u63a5\u4ece encoder \u7684\u8f93\u51fa\u63d0\u53d6 features \u548c reference boxes\uff09\uff1b\u5176\u6b21\u662f\u5b83\u7528\u4e86\u6240\u8c13\u68af\u5ea6\u5206\u79bb\u6280\u672f\uff0c\u8fd9\u91cc\u53eb\u5b83 look forward once<\/li>\n<\/ul>\n<h2 id=\"backbone\">Backbone<\/h2>\n<p>\u5206\u522b\u7528 ResNet-50 \u548c SwinL \u4f5c\u4e3a backbone<\/p>\n<h2 id=\"\u6a21\u578b\">\u6a21\u578b<\/h2>\n<ul>\n<li>look forward twice: \u4e00\u4e2a\u7c7b\u4f3c residual connection \u7684\u64cd\u4f5c\uff0c\u7528\u6765\u51cf\u5c11\u68af\u5ea6\u6d88\u5931\u7684\u98ce\u9669<\/li>\n<li>Mixed Query Selection: \u9009\u62e9 top-k \u4e2a feature \u4f20\u5165 decoder<\/li>\n<\/ul>\n<h2 id=\"\u8bad\u7ec3-1\">\u8bad\u7ec3<\/h2>\n<ul>\n<li>Contrastive DeNoising Training (CDN) \uff0c\u5c06 Ground Truth \u5185\u7684\u4f5c\u4e3a\u6b63\u6837\u672c\uff0c\u5916\u7684\u4f5c\u4e3a\u8d1f\u6837\u672c\uff0c\u53ef\u4ee5\u907f\u514d\u91cd\u590d\u9884\u6d4b\u3002<\/li>\n<\/ul>\n<h1 id=\"grounding-dino-\u9605\u8bfb\u7b14\u8bb0\">Grounding DINO \u9605\u8bfb\u7b14\u8bb0<\/h1>\n<h2 id=\"preliminaries-1\">Preliminaries<\/h2>\n<p>\u63d0\u51fa\u4e00\u4e2a\u8303\u5f0f\uff1aBackbone \u7528\u4e8e feature extraction, neck \u7528\u4e8e feature enchancement, head \u7528\u4e8e region refinement(box prediction)<\/p>\n<h2 id=\"\u6a21\u578b-1\">\u6a21\u578b<\/h2>\n<p>image \u7ecf\u8fc7\u4e00\u4e2a Swin-T \u4f5c\u4e3a backbone, \u800c text \u5219\u7ecf\u8fc7\u4e00\u4e2a BERT<\/p>\n<p>\u7136\u540e\u662f\u4e00\u4e2a Feature Enhance Layer, SAM \u548c\u5b83\u6709\u70b9\u76f8\u4f3c\uff0c\u90fd\u662f image \u548c text (\u5728 SAM \u91cc\u662f prompt) \u5404\u7ecf\u8fc7\u4e00\u4e2a self-attention \/ deformable self-attention \u4ee5\u540e\u505a\u4e00\u4e2a\u53cc\u5411\u7684 cross-attention, \u7136\u540e\u5404 FFN \u4ee5\u540e\u8f93\u51fa\u3002<\/p>\n<p>\u63a5\u7740\u662f Language-Guided Query Selection, \u8fd9\u4e2a\u90e8\u5206\u9996\u5148\u5c06 image \u548c text \u7684 feature \u6cbf\u7b2c\u4e09\u7ef4\u5377\u79ef\uff0c\u53d6 max \u4f5c\u4e3a img feat\uff0c\u6700\u540e\u53d6 topk \u8fd4\u56de\u3002<\/p>\n<p>\u63a5\u7740\u662f Cross-Modality Decoder, \u5c06 Query \u7ecf\u8fc7 self-attention \u4ee5\u540e\uff0c\u53d6 Q \u5206\u522b\u548c Image Features, Text Features \u7684 K, V \u505a Cross-Attention \uff0c\u518d\u8fc7\u4e00\u4e2a FFN \u4ee5\u540e\u8f93\u51fa\u3002<\/p>\n<h2 id=\"\u8f93\u5165\">\u8f93\u5165<\/h2>\n<p>\u5904\u7406 Text Prompt: \u4ee5\u5f80\u91c7\u7528\u4e24\u79cd text prompt, \u4f5c\u8005\u79f0\u4e4b\u4e3a Sentence Level \u548c Word Level; \u5728\u8fd9\u91cc\u5219\u4f7f\u7528\u4e86 Sub-sentence Level<\/p>\n<h2 id=\"\u8bad\u7ec3-2\">\u8bad\u7ec3<\/h2>\n<ul>\n<li>\u7c7b\u4f3c DETR, \u4ecd\u7136\u662f\u5148\u4e8c\u5206\u56fe\u5339\u914d\u518d\u8ba1\u7b97\u635f\u5931<\/li>\n<\/ul>\n<h2 id=\"ablations\">Ablations<\/h2>\n<p>\u6d88\u878d\u5b9e\u9a8c\u8ba4\u4e3a Encoder fusion \u662f\u6700\u91cd\u8981\u7684\u90e8\u4efd<\/p>\n\n \n \n <\/body>\n <\/html>\n","protected":false},"excerpt":{"rendered":"<p>DEtection with TRansformers 阅读笔&#x … <\/p>\n<p class=\"link-more\"><a href=\"http:\/\/SmokeyDays.top\/wordpress\/2023\/07\/15\/grounding-dino-%e9%98%85%e8%af%bb%e7%ac%94%e8%ae%b0\/\" class=\"more-link\">\u7ee7\u7eed\u9605\u8bfb<span class=\"screen-reader-text\">\u201c(grounding) DINO \u9605\u8bfb\u7b14\u8bb0\u201d<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[131,130,135],"tags":[],"_links":{"self":[{"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/posts\/1074"}],"collection":[{"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/comments?post=1074"}],"version-history":[{"count":1,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/posts\/1074\/revisions"}],"predecessor-version":[{"id":1075,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/posts\/1074\/revisions\/1075"}],"wp:attachment":[{"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/media?parent=1074"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/categories?post=1074"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/SmokeyDays.top\/wordpress\/wp-json\/wp\/v2\/tags?post=1074"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}