{"id":3777,"date":"2024-10-30T14:45:04","date_gmt":"2024-10-30T14:45:04","guid":{"rendered":"https:\/\/salarydistribution.com\/machine-learning\/2024\/10\/30\/spooks-await-at-the-haunted-sanctuary-built-with-rtx-and-ai\/"},"modified":"2024-10-30T14:45:04","modified_gmt":"2024-10-30T14:45:04","slug":"spooks-await-at-the-haunted-sanctuary-built-with-rtx-and-ai","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2024\/10\/30\/spooks-await-at-the-haunted-sanctuary-built-with-rtx-and-ai\/","title":{"rendered":"Spooks Await at the \u2018Haunted Sanctuary,\u2019 Built With RTX and AI"},"content":{"rendered":"<div>\n\t\t<span class=\"bsf-rt-reading-time\"><span class=\"bsf-rt-display-label\"><\/span> <span class=\"bsf-rt-display-time\"><\/span> <span class=\"bsf-rt-display-postfix\"><\/span><\/span><\/p>\n<p>Among the artists using AI to enhance and accelerate their creative endeavors is Sabour Amirazodi, a creator and tech marketing and workflow specialist at NVIDIA.<\/p>\n<p>Using his over 20 years of multi-platform experience in location-based entertainment and media production, he decorates his home every year with an incredible Halloween installation \u2014 dubbed the <i>Haunted Sanctuary<\/i>.<\/p>\n<\/p>\n<p>The project is a massive undertaking requiring projection mapping, the creation and assembly of 3D scenes, compositing and editing in Adobe After Effects and Premiere Pro, and more. The creation process was accelerated using the <a target=\"_blank\" href=\"https:\/\/www.nvidia.com\/en-us\/studio\/\" rel=\"noopener\">NVIDIA Studio<\/a> content creation platform and Amirazodi\u2019s <a target=\"_blank\" href=\"https:\/\/www.nvidia.com\/en-us\/design-visualization\/rtx\/\" rel=\"noopener\">NVIDIA RTX 6000 GPU<\/a>.<\/p>\n<p>This year, Amirazodi deployed new AI workflows in ComfyUI, Adobe Firefly and Photoshop to create digital portraits \u2014 inspired by his family \u2014 as part of the installation.<\/p>\n<h2><b>Give \u2019em Pumpkin to Talk About<\/b><\/h2>\n<p><a target=\"_blank\" href=\"https:\/\/www.comfy.org\/\" rel=\"noopener\">ComfyUI<\/a> is a node-based interface that generates images and videos from text. It\u2019s designed to be highly customizable, allowing users to design workflows, adjust settings and see results immediately. It can combine various AI models and third-party extensions to achieve a higher degree of control.<\/p>\n<p>For example, this workflow below requires entering a prompt, the details and characteristics of the desired image, and a negative prompt to help omit any undesired visual effects.<\/p>\n<p><a href=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/CartoonFamily_AI.gif\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full wp-image-75068\" src=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/CartoonFamily_AI.gif\" alt=\"\" width=\"1028\" height=\"578\"><\/a><\/p>\n<p>Since Amirazodi wanted his digital creations to closely resemble his family, he started by applying Run IP Adapters, which use reference images to inform generated content.<\/p>\n<figure id=\"attachment_75071\" aria-describedby=\"caption-attachment-75071\" class=\"wp-caption aligncenter\"><a href=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/comfyUI.png\"><img decoding=\"async\" loading=\"lazy\" class=\"size-large wp-image-75071\" src=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/comfyUI-1680x819.png\" alt=\"\" width=\"1680\" height=\"819\"><\/a><figcaption id=\"caption-attachment-75071\" class=\"wp-caption-text\">ComfyUI nodes and reference material in the viewer.<\/figcaption><\/figure>\n<p>From there, he tinkered with the settings to achieve the desired look and feel of each character.<\/p>\n<figure id=\"attachment_75074\" aria-describedby=\"caption-attachment-75074\" class=\"wp-caption aligncenter\"><a href=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/The-Amirazodis.png\"><img decoding=\"async\" loading=\"lazy\" class=\"size-full wp-image-75074\" src=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/The-Amirazodis.png\" alt=\"\" width=\"1105\" height=\"809\"><\/a><figcaption id=\"caption-attachment-75074\" class=\"wp-caption-text\">The Amirazodis digitized for the \u2018Halloween Sanctuary\u2019 installation.<\/figcaption><\/figure>\n<p>ComfyUI has <a target=\"_blank\" href=\"https:\/\/github.com\/comfyanonymous\/ComfyUI_TensorRT\" rel=\"noopener\">NVIDIA TensorRT acceleration<\/a>, so RTX users can generate images from prompts up to 60% faster.<\/p>\n<\/p>\n<p>Get started with <a target=\"_blank\" href=\"https:\/\/www.comfy.org\/\" rel=\"noopener\">ComfyUI<\/a>.<\/p>\n<h2><b>In Darkness, Let There Be Light<\/b><\/h2>\n<p>Adobe Firefly is a family of creative generative AI models that offer new ways to ideate and create while assisting creative workflows. They\u2019re designed to be safe for commercial use and were trained, using NVIDIA GPUs, on licensed content like Adobe Stock Images and public domain content where copyright has expired.<\/p>\n<p>To make the digital portraits fit as desired, Amirazodi needed to expand the background.<\/p>\n<p>Adobe Photoshop features a Generative Fill tool called Generative Expand that allows artists to extend the border of their image with the Crop tool and automatically fill the space with content that matches the existing image.<\/p>\n<p><a href=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/Adobe_Firefly.gif\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full wp-image-75077\" src=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/Adobe_Firefly.gif\" alt=\"\" width=\"772\" height=\"434\"><\/a><\/p>\n<p>Photoshop also features \u201cNeural Filters that allow artists to explore creative ideas and make complex adjustments to images in just seconds, saving them hours of tedious, manual work.<\/p>\n<\/p>\n<p>With Smart Portrait Neural Filters, artists can easily experiment with facial characteristics such as gaze direction and lighting angles simply by dragging a slider. Amirazodi used the feature to apply the final touches to his portraits, adjusting colors, textures, depth blur and facial expressions.<\/p>\n<p><a href=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/Adobe_filters.gif\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full wp-image-75080\" src=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/Adobe_filters.gif\" alt=\"\" width=\"748\" height=\"420\"><\/a><\/p>\n<p>NVIDIA RTX GPUs help power AI-based tasks, accelerating the Neural Filters in Photoshop.<\/p>\n<p><a href=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/CartoonFamily_AI_ComfyUI-Firefly.gif\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full wp-image-75083\" src=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/CartoonFamily_AI_ComfyUI-Firefly.gif\" alt=\"\" width=\"882\" height=\"496\"><\/a><\/p>\n<p>Learn more about the latest Adobe features and tools in this <a href=\"https:\/\/blogs.nvidia.com\/blog\/studio-adobe-max-firefly-ai-substance-3d-viewer\/\">blog<\/a>.<\/p>\n<p><a href=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/001_HS_2024_Garage_SHOW.gif\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full wp-image-75139\" src=\"https:\/\/blogs.nvidia.com\/wp-content\/uploads\/2024\/10\/001_HS_2024_Garage_SHOW.gif\" alt=\"\" width=\"1012\" height=\"570\"><\/a><\/p>\n<p>AI is already helping accelerate and automate tasks across content creation, gaming and everyday life \u2014 and the speedups are only multiplied with an NVIDIA RTX- or GeForce RTX GPU-equipped system.<\/p>\n<p>Check out and share Halloween- and fall-themed art as a part of the NVIDIA Studio #HarvestofCreativity challenge on <a target=\"_blank\" href=\"https:\/\/www.instagram.com\/nvidiastudio\/\" rel=\"noopener\">Instagram<\/a>, <a target=\"_blank\" href=\"https:\/\/x.com\/NVIDIAStudio\" rel=\"noopener\">X<\/a>, <a target=\"_blank\" href=\"https:\/\/www.facebook.com\/NVIDIAStudio\/\" rel=\"noopener\">Facebook<\/a> and <a target=\"_blank\" href=\"https:\/\/www.threads.net\/@nvidiastudio\" rel=\"noopener\">Threads<\/a> for a chance to be featured on the social media channels.<\/p>\n<\/p><\/div>\n","protected":false},"excerpt":{"rendered":"<p>https:\/\/blogs.nvidia.com\/blog\/ai-studio-comfyui-adobe-firefly-photoshop\/<\/p>\n","protected":false},"author":0,"featured_media":3778,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[3],"tags":[],"_links":{"self":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/3777"}],"collection":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/comments?post=3777"}],"version-history":[{"count":0,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/3777\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media\/3778"}],"wp:attachment":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media?parent=3777"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/categories?post=3777"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/tags?post=3777"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}