{"id":1528,"date":"2023-09-13T12:02:31","date_gmt":"2023-09-13T12:02:31","guid":{"rendered":"https:\/\/gyrus.ai\/blog\/?p=1528"},"modified":"2023-09-14T10:00:06","modified_gmt":"2023-09-14T10:00:06","slug":"comparison-realtime-vision-ai-pipelines","status":"publish","type":"post","link":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/","title":{"rendered":"A Comparison of Real Time Vision AI Pipelines"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Introduction<\/span><\/h1>

This article is based on our expertise in building AI vision models and deploying them in customized real time video processing pipelines<\/a><\/strong>. In our experience,a realtime vision Ai pipeline enables fast results and saves on GPU hardware usage cost.<\/span><\/p>

\u00a0<\/span><\/p>

This is a comparative study between two popular frameworks, which we have customized and deployed extensively in our Vision Ai based products.For this study, we have used an application which detects multiple faces using a face detection model<\/a> and creates face masks to precisely blur face contours. Our aim is to maximize the throughput and achieve real time frame processing speeds.<\/span><\/p>

\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0<\/h3>

AI Vision Solution Frameworks<\/h3>
\u00a0<\/div>

An AI vision solution requires pre and post processing modules apart from the Vision model.\u00a0 A solution architect has to select an appropriate framework for deploying an end to end pipeline with all the processing stages and AI model. <\/span><\/p>

\u00a0<\/p><\/div>

\"Components\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Components of a AI Video Processing Pipeline<\/span><\/p>

\u00a0<\/span><\/p>

A single video stream<\/a> generates around 25 – 30 frames per second. Vision AI\u00a0 inference pipeline<\/strong> has to process these frames\/images with minimum latency.\u00a0 \u00a0<\/span>For real time performance, the pipeline has to process input frames within the input frame rate. Throughput is measured in FPS( frames processed per second). We will explore pros and cons of two broad deployment strategies, Deepstream\/ Gstreamer\/ C++ plugin based pipeline Vs. Python\/Pytorch inference script.<\/span><\/p><\/div>

Python\/Pytorch<\/i><\/b> based inference script is a popular approach during the proof of concept stage of a solution. Any vision Ai model<\/a> can easily be used in inference mode with few lines of python code. Pytorch\/Opencv provides most of the frame transform operations needed for pre\/post processing of frames. <\/span>Generally python\/pytorch based solutions are slow in processing frames and can\u2019t be scaled easily<\/span><\/i>.\u00a0<\/span><\/p>

Overall this approach contains one main loop which contains step by step processing functions along with model inference code. <\/span>It doesn\u2019t become a true pipeline where multiple stages can work in parallel fashion to accomplish overall tasks faster.\u00a0<\/span><\/i><\/p>

A good example of a pipeline is an automobile assembly line which has multiple independent stages and at any given time multiple cars will be going through this pipeline. This is essential to achieve efficiency.<\/span><\/p>

Deepstream<\/i><\/b> is a framework from NVIDIA which is used to deploy AI vision models<\/a> on Nvidia\u2019s GPU hardware. It has inbuilt plugins for various tasks required for model inference and frame processing.\u00a0<\/span><\/p>

Plugins work as independent pipeline stages which can process frames in parallel. This framework promises faster performance based on true pipeline architecture.<\/span><\/p>

\"\"\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0A Deepstream Pipeline For AI Video Analytics<\/span><\/p><\/div>

These plugins can be customized for different requirements or new ones can be developed from scratch.\u00a0<\/span><\/p>

Nvinfer is the main inference plugin in Deepstream framework. By default, It supports object detection models in Onnx or TensorRT format for running inference.<\/span><\/p>

For other computer vision tasks, this plugin needs to be customized based on the model\u2019s input and output requirements. <\/span>We have developed a customized\u00a0version of Nvinfer plugin for running our version of face detection model and blurring out detected faces in each frame.<\/span><\/p>

Performance Comparison:<\/b><\/h3>

\"Performance<\/p><\/div>

Deepstream pipeline<\/strong><\/em> requires 15% less GPU memory for end to end execution. Internally deepstream uses NVIDIA TensorRT runtime, which claims to optimize GPU memory usage <\/span>by fusing model<\/span> nodes and reusing tensor memory efficiently. <\/span><\/p>

\"Frames
<\/span><\/p>

Demo Video<\/span><\/h3>
\nhttps:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/08\/Untitled-design.mp4<\/a><\/video><\/div>

Original video Source and license –\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 <\/span>https:\/\/github.com\/intel-iot-devkit\/sample-videos\/tree\/master<\/a><\/span><\/p>

Conclusion\u00a0<\/h3><\/div>
\u00a0<\/div>

Deepstream pipeline is 9x faster in processing frames, it is able to process 65 frames\/sec and far exceeds realtime Fps requirement.\u00a0 This speedup is due to parallel execution of multiple frames in the pipeline and optimized model inference runtime.\u00a0<\/span><\/p>

A python\/pytorch pipeline will not be efficient for video frame processing<\/a> in realtime, it will require more memory and faster expensive Gpus to increase frame processing speed.<\/span><\/p>

(Deepstream and Nvinfer are Nvidia\u2019s trademark)<\/span><\/p><\/div><\/div>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"

Introduction This article is based on our expertise in building AI vision models and deploying them … A Comparison of Real Time Vision AI Pipelines<\/span>Read more<\/a><\/p>\n","protected":false},"author":9,"featured_media":1767,"comment_status":"open","ping_status":"open","sticky":false,"template":"elementor_theme","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[6],"tags":[95,98,96,97,99],"class_list":["post-1528","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-technology","tag-ai-in-video-stream","tag-ai-video-analytics","tag-ai-vision-models","tag-deepstream","tag-video-frame-processing"],"yoast_head":"\nA Comparison of Real Time Vision AI Pipelines<\/title>\n<meta name=\"description\" content=\"An Real Time AI vision pipelines requires pre and post processing modules apart from the AI Vision model. Deepstream is a Video stream and video redactor framework from NVIDIA which is used to deploy AI vision models on Nvidia\u2019s GPU hardware.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"A Comparison of Real Time Vision AI Pipelines\" \/>\n<meta property=\"og:description\" content=\"An Real Time AI vision pipelines requires pre and post processing modules apart from the AI Vision model. Deepstream is a Video stream and video redactor framework from NVIDIA which is used to deploy AI vision models on Nvidia\u2019s GPU hardware.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/\" \/>\n<meta property=\"og:site_name\" content=\"Gyrus AI | Blog | Insights on AI & Intelligent Media Search, In-scene Ad Placement, Automated Video Anonymization Technologies\" \/>\n<meta property=\"article:published_time\" content=\"2023-09-13T12:02:31+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-09-14T10:00:06+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2-1024x662.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1024\" \/>\n\t<meta property=\"og:image:height\" content=\"662\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Abhishek Singh\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@GyrusAI\" \/>\n<meta name=\"twitter:site\" content=\"@GyrusAI\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Abhishek Singh\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/\"},\"author\":{\"name\":\"Abhishek Singh\",\"@id\":\"https:\/\/gyrus.ai\/blog\/#\/schema\/person\/95e2933bc15d9ec9bec478c1587d7ede\"},\"headline\":\"A Comparison of Real Time Vision AI Pipelines\",\"datePublished\":\"2023-09-13T12:02:31+00:00\",\"dateModified\":\"2023-09-14T10:00:06+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/\"},\"wordCount\":657,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/#organization\"},\"image\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2.png\",\"keywords\":[\"AI in Video Stream\",\"AI Video Analytics\",\"AI vision models\",\"Deepstream\",\"video frame processing\"],\"articleSection\":[\"Technology\"],\"inLanguage\":\"en\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/\",\"url\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/\",\"name\":\"A Comparison of Real Time Vision AI Pipelines\",\"isPartOf\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2.png\",\"datePublished\":\"2023-09-13T12:02:31+00:00\",\"dateModified\":\"2023-09-14T10:00:06+00:00\",\"description\":\"An Real Time AI vision pipelines requires pre and post processing modules apart from the AI Vision model. Deepstream is a Video stream and video redactor framework from NVIDIA which is used to deploy AI vision models on Nvidia\u2019s GPU hardware.\",\"breadcrumb\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#breadcrumb\"},\"inLanguage\":\"en\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en\",\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#primaryimage\",\"url\":\"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2.png\",\"contentUrl\":\"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2.png\",\"width\":2560,\"height\":1656,\"caption\":\"Deepstream Vs. Pytorch inference\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/gyrus.ai\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"A Comparison of Real Time Vision AI Pipelines\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/gyrus.ai\/blog\/#website\",\"url\":\"https:\/\/gyrus.ai\/blog\/\",\"name\":\"Gyrus AI\",\"description\":\"Gyrus AI | Blog | Insights on AI & Intelligent Media Search, In-scene Ad Placement, Automated Video Anonymization Technologies\",\"publisher\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/gyrus.ai\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/gyrus.ai\/blog\/#organization\",\"name\":\"Gyrus AI\",\"url\":\"https:\/\/gyrus.ai\/blog\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en\",\"@id\":\"https:\/\/gyrus.ai\/blog\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2025\/02\/gyrus-1.png\",\"contentUrl\":\"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2025\/02\/gyrus-1.png\",\"width\":11400,\"height\":4500,\"caption\":\"Gyrus AI\"},\"image\":{\"@id\":\"https:\/\/gyrus.ai\/blog\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/x.com\/GyrusAI\",\"https:\/\/www.linkedin.com\/company\/gyrusai\/\",\"https:\/\/www.youtube.com\/channel\/UCk2GzLj6xp0A6Wqix1GWSkw\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/gyrus.ai\/blog\/#\/schema\/person\/95e2933bc15d9ec9bec478c1587d7ede\",\"name\":\"Abhishek Singh\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en\",\"@id\":\"https:\/\/secure.gravatar.com\/avatar\/5fdf8a1251d0f3487bdc5ed81c91b93e1c8bf7027e491018f6fb6fd7871e5378?s=96&d=mm&r=g\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/5fdf8a1251d0f3487bdc5ed81c91b93e1c8bf7027e491018f6fb6fd7871e5378?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/5fdf8a1251d0f3487bdc5ed81c91b93e1c8bf7027e491018f6fb6fd7871e5378?s=96&d=mm&r=g\",\"caption\":\"Abhishek Singh\"},\"url\":\"https:\/\/gyrus.ai\/blog\/author\/abhishek-singh\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"A Comparison of Real Time Vision AI Pipelines","description":"An Real Time AI vision pipelines requires pre and post processing modules apart from the AI Vision model. Deepstream is a Video stream and video redactor framework from NVIDIA which is used to deploy AI vision models on Nvidia\u2019s GPU hardware.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/","og_locale":"en_US","og_type":"article","og_title":"A Comparison of Real Time Vision AI Pipelines","og_description":"An Real Time AI vision pipelines requires pre and post processing modules apart from the AI Vision model. Deepstream is a Video stream and video redactor framework from NVIDIA which is used to deploy AI vision models on Nvidia\u2019s GPU hardware.","og_url":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/","og_site_name":"Gyrus AI | Blog | Insights on AI & Intelligent Media Search, In-scene Ad Placement, Automated Video Anonymization Technologies","article_published_time":"2023-09-13T12:02:31+00:00","article_modified_time":"2023-09-14T10:00:06+00:00","og_image":[{"width":1024,"height":662,"url":"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2-1024x662.png","type":"image\/png"}],"author":"Abhishek Singh","twitter_card":"summary_large_image","twitter_creator":"@GyrusAI","twitter_site":"@GyrusAI","twitter_misc":{"Written by":"Abhishek Singh","Est. reading time":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#article","isPartOf":{"@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/"},"author":{"name":"Abhishek Singh","@id":"https:\/\/gyrus.ai\/blog\/#\/schema\/person\/95e2933bc15d9ec9bec478c1587d7ede"},"headline":"A Comparison of Real Time Vision AI Pipelines","datePublished":"2023-09-13T12:02:31+00:00","dateModified":"2023-09-14T10:00:06+00:00","mainEntityOfPage":{"@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/"},"wordCount":657,"commentCount":0,"publisher":{"@id":"https:\/\/gyrus.ai\/blog\/#organization"},"image":{"@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#primaryimage"},"thumbnailUrl":"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2.png","keywords":["AI in Video Stream","AI Video Analytics","AI vision models","Deepstream","video frame processing"],"articleSection":["Technology"],"inLanguage":"en","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/","url":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/","name":"A Comparison of Real Time Vision AI Pipelines","isPartOf":{"@id":"https:\/\/gyrus.ai\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#primaryimage"},"image":{"@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#primaryimage"},"thumbnailUrl":"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2.png","datePublished":"2023-09-13T12:02:31+00:00","dateModified":"2023-09-14T10:00:06+00:00","description":"An Real Time AI vision pipelines requires pre and post processing modules apart from the AI Vision model. Deepstream is a Video stream and video redactor framework from NVIDIA which is used to deploy AI vision models on Nvidia\u2019s GPU hardware.","breadcrumb":{"@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#breadcrumb"},"inLanguage":"en","potentialAction":[{"@type":"ReadAction","target":["https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/"]}]},{"@type":"ImageObject","inLanguage":"en","@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#primaryimage","url":"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2.png","contentUrl":"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2023\/09\/Deepstream-Vs.-Pytorch-inference-2.png","width":2560,"height":1656,"caption":"Deepstream Vs. Pytorch inference"},{"@type":"BreadcrumbList","@id":"https:\/\/gyrus.ai\/blog\/comparison-realtime-vision-ai-pipelines\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/gyrus.ai\/blog\/"},{"@type":"ListItem","position":2,"name":"A Comparison of Real Time Vision AI Pipelines"}]},{"@type":"WebSite","@id":"https:\/\/gyrus.ai\/blog\/#website","url":"https:\/\/gyrus.ai\/blog\/","name":"Gyrus AI","description":"Gyrus AI | Blog | Insights on AI & Intelligent Media Search, In-scene Ad Placement, Automated Video Anonymization Technologies","publisher":{"@id":"https:\/\/gyrus.ai\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/gyrus.ai\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en"},{"@type":"Organization","@id":"https:\/\/gyrus.ai\/blog\/#organization","name":"Gyrus AI","url":"https:\/\/gyrus.ai\/blog\/","logo":{"@type":"ImageObject","inLanguage":"en","@id":"https:\/\/gyrus.ai\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2025\/02\/gyrus-1.png","contentUrl":"https:\/\/gyrus.ai\/blog\/wp-content\/uploads\/2025\/02\/gyrus-1.png","width":11400,"height":4500,"caption":"Gyrus AI"},"image":{"@id":"https:\/\/gyrus.ai\/blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/GyrusAI","https:\/\/www.linkedin.com\/company\/gyrusai\/","https:\/\/www.youtube.com\/channel\/UCk2GzLj6xp0A6Wqix1GWSkw"]},{"@type":"Person","@id":"https:\/\/gyrus.ai\/blog\/#\/schema\/person\/95e2933bc15d9ec9bec478c1587d7ede","name":"Abhishek Singh","image":{"@type":"ImageObject","inLanguage":"en","@id":"https:\/\/secure.gravatar.com\/avatar\/5fdf8a1251d0f3487bdc5ed81c91b93e1c8bf7027e491018f6fb6fd7871e5378?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/5fdf8a1251d0f3487bdc5ed81c91b93e1c8bf7027e491018f6fb6fd7871e5378?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/5fdf8a1251d0f3487bdc5ed81c91b93e1c8bf7027e491018f6fb6fd7871e5378?s=96&d=mm&r=g","caption":"Abhishek Singh"},"url":"https:\/\/gyrus.ai\/blog\/author\/abhishek-singh\/"}]}},"_links":{"self":[{"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/posts\/1528","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/users\/9"}],"replies":[{"embeddable":true,"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/comments?post=1528"}],"version-history":[{"count":192,"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/posts\/1528\/revisions"}],"predecessor-version":[{"id":1766,"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/posts\/1528\/revisions\/1766"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/media\/1767"}],"wp:attachment":[{"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/media?parent=1528"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/categories?post=1528"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/gyrus.ai\/blog\/wp-json\/wp\/v2\/tags?post=1528"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}