{"id":97,"date":"2025-12-12T04:43:49","date_gmt":"2025-12-12T04:43:49","guid":{"rendered":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/?page_id=97"},"modified":"2025-12-12T21:17:17","modified_gmt":"2025-12-12T21:17:17","slug":"experiments","status":"publish","type":"page","link":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/","title":{"rendered":"Experiments"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\">Visual Comparisons<\/h2>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"954\" height=\"1024\" src=\"http:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-954x1024.jpg\" alt=\"\" class=\"wp-image-114\" srcset=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-954x1024.jpg 954w, https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-280x300.jpg 280w, https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-768x824.jpg 768w, https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-1432x1536.jpg 1432w, https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-1909x2048.jpg 1909w\" sizes=\"auto, (max-width: 767px) 89vw, (max-width: 1000px) 54vw, (max-width: 1071px) 543px, 580px\" \/><\/figure>\n\n\n\n<p>We compare our approach with other enhancement methods. <strong>SuperHead<\/strong> synthesizes high-quality facial details across diverse expressions, clearly outperforming baselines and in some cases approaching the pseudo ground-truth head avatar. All methods are driven and rendered with novel camera poses and expressions.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Quantitative Results<\/h2>\n\n\n\n<p><strong>State-of-the-art Performance<\/strong>: We evaluated SuperHead on the NeRSemble and INSTA<strong> <\/strong>datasets. Our method consistently achieves the best scores across key image quality metrics:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>PSNR \/ SSIM (<math data-latex=\"\\uparrow\"><semantics><mo stretchy=\"false\" lspace=\"0em\" rspace=\"0em\">\u2191<\/mo><annotation encoding=\"application\/x-tex\">\\uparrow<\/annotation><\/semantics><\/math>)<\/strong>: Higher values indicate better fidelity. SuperHead outperforms all baselines.<\/li>\n\n\n\n<li><strong>LPIPS (<\/strong><math data-latex=\"\\downarrow\"><semantics><mo stretchy=\"false\" lspace=\"0em\" rspace=\"0em\">\u2193<\/mo><annotation encoding=\"application\/x-tex\">\\downarrow<\/annotation><\/semantics><\/math><strong>)<\/strong>: Lower values indicate better perceptual quality. Our method produces the most natural-looking results.<\/li>\n\n\n\n<li><strong>Efficient Processing<\/strong>: Thanks to our efficient 3D GAN inversion, SuperHead requires significantly less inference time compared to video-based super-resolution methods (e.g., <strong>5 mins<\/strong> vs. 30 mins for a sequence), making it a practical solution for scalable avatar creation.<\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"254\" src=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/image-1-1024x254.png\" alt=\"\" class=\"wp-image-125\" srcset=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/image-1-1024x254.png 1024w, https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/image-1-300x74.png 300w, https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/image-1-768x191.png 768w, https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/image-1.png 1451w\" sizes=\"auto, (max-width: 767px) 89vw, (max-width: 1000px) 54vw, (max-width: 1071px) 543px, 580px\" \/><\/figure>\n\n\n\n<p>For more video comparisons, see the <a href=\"https:\/\/willydjhuang.github.io\/superhead\/\">SuperHead website<\/a> for more information.<\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Visual Comparisons We compare our approach with other enhancement methods. SuperHead synthesizes high-quality facial details across diverse expressions, clearly outperforming baselines and in some cases approaching the pseudo ground-truth head avatar. All methods are driven and rendered with novel camera poses and expressions. Quantitative Results State-of-the-art Performance: We evaluated SuperHead on the NeRSemble and INSTA &hellip; <\/p>\n<p class=\"link-more\"><a href=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Experiments&#8221;<\/span><\/a><\/p>\n","protected":false},"author":259,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-97","page","type-page","status-publish","hentry"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.1 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Experiments - 3D Gaussian Human Enhancement<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Experiments - 3D Gaussian Human Enhancement\" \/>\n<meta property=\"og:description\" content=\"Visual Comparisons We compare our approach with other enhancement methods. SuperHead synthesizes high-quality facial details across diverse expressions, clearly outperforming baselines and in some cases approaching the pseudo ground-truth head avatar. All methods are driven and rendered with novel camera poses and expressions. Quantitative Results State-of-the-art Performance: We evaluated SuperHead on the NeRSemble and INSTA &hellip; Continue reading &quot;Experiments&quot;\" \/>\n<meta property=\"og:url\" content=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/\" \/>\n<meta property=\"og:site_name\" content=\"3D Gaussian Human Enhancement\" \/>\n<meta property=\"article:modified_time\" content=\"2025-12-12T21:17:17+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-scaled.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"2386\" \/>\n\t<meta property=\"og:image:height\" content=\"2560\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/experiments\\\/\",\"url\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/experiments\\\/\",\"name\":\"Experiments - 3D Gaussian Human Enhancement\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/experiments\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/experiments\\\/#primaryimage\"},\"thumbnailUrl\":\"http:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/wp-content\\\/uploads\\\/sites\\\/137\\\/2025\\\/12\\\/result-954x1024.jpg\",\"datePublished\":\"2025-12-12T04:43:49+00:00\",\"dateModified\":\"2025-12-12T21:17:17+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/experiments\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/experiments\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/experiments\\\/#primaryimage\",\"url\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/wp-content\\\/uploads\\\/sites\\\/137\\\/2025\\\/12\\\/result-scaled.jpg\",\"contentUrl\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/wp-content\\\/uploads\\\/sites\\\/137\\\/2025\\\/12\\\/result-scaled.jpg\",\"width\":2386,\"height\":2560},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/experiments\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Experiments\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/#website\",\"url\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/\",\"name\":\"3D Gaussian Human Enhancement\",\"description\":\"2025 MSCV Capstone Project\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2025team11\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Experiments - 3D Gaussian Human Enhancement","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:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/","og_locale":"en_US","og_type":"article","og_title":"Experiments - 3D Gaussian Human Enhancement","og_description":"Visual Comparisons We compare our approach with other enhancement methods. SuperHead synthesizes high-quality facial details across diverse expressions, clearly outperforming baselines and in some cases approaching the pseudo ground-truth head avatar. All methods are driven and rendered with novel camera poses and expressions. Quantitative Results State-of-the-art Performance: We evaluated SuperHead on the NeRSemble and INSTA &hellip; Continue reading \"Experiments\"","og_url":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/","og_site_name":"3D Gaussian Human Enhancement","article_modified_time":"2025-12-12T21:17:17+00:00","og_image":[{"width":2386,"height":2560,"url":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-scaled.jpg","type":"image\/jpeg"}],"twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/","url":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/","name":"Experiments - 3D Gaussian Human Enhancement","isPartOf":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/#website"},"primaryImageOfPage":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/#primaryimage"},"image":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/#primaryimage"},"thumbnailUrl":"http:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-954x1024.jpg","datePublished":"2025-12-12T04:43:49+00:00","dateModified":"2025-12-12T21:17:17+00:00","breadcrumb":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/#primaryimage","url":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-scaled.jpg","contentUrl":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-content\/uploads\/sites\/137\/2025\/12\/result-scaled.jpg","width":2386,"height":2560},{"@type":"BreadcrumbList","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/experiments\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/"},{"@type":"ListItem","position":2,"name":"Experiments"}]},{"@type":"WebSite","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/#website","url":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/","name":"3D Gaussian Human Enhancement","description":"2025 MSCV Capstone Project","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"}]}},"_links":{"self":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-json\/wp\/v2\/pages\/97","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-json\/wp\/v2\/users\/259"}],"replies":[{"embeddable":true,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-json\/wp\/v2\/comments?post=97"}],"version-history":[{"count":4,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-json\/wp\/v2\/pages\/97\/revisions"}],"predecessor-version":[{"id":127,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-json\/wp\/v2\/pages\/97\/revisions\/127"}],"wp:attachment":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2025team11\/wp-json\/wp\/v2\/media?parent=97"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}