{"id":19,"date":"2022-04-29T01:44:05","date_gmt":"2022-04-29T01:44:05","guid":{"rendered":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/?page_id=19"},"modified":"2022-12-21T04:48:17","modified_gmt":"2022-12-21T04:48:17","slug":"results","status":"publish","type":"page","link":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/","title":{"rendered":"Results"},"content":{"rendered":"\n

Summary – We achieve SoTA in MOT20 and 2nd place in MOT17 on the official MOT Challenge Leaderboard <\/strong><\/mark><\/p>\n\n\n\n

Our failure analysis on SoTA algorithms such as OC-SORT led us to establish that occlusion-based false negatives and ID swaps are the biggest source of error for SORT based trackers. Additionally, we found that naively adding appearance modeling actually reduces performance. <\/p>\n\n\n\n

To counteract these we introduce Grid Based Appearance modeling, providing a more fine-grained embedding compared to typical bounding box level descriptions. We further add Dynamic Appearance, Adaptive Weighting, and CMC to OC-SORT. With the the help of all these changes we are able to attain SoTA on MOT20<\/strong> and 2nd on MOT17<\/strong> on the official MOT Challenge Leaderboard. <\/p>\n\n\n\n

Ablation Study – Grid-based Appearance Modeling and Hard Thresholds<\/h2>\n\n\n\n
\"\"
Grid-Based Appearance Model – Results using different grid splits (without post-processing). We find that a 3×1 horizontal split works the best<\/figcaption><\/figure><\/div>\n\n\n\n
\"\"
Hard Thresholds – Results (without post-processing) for different threshold values.<\/figcaption><\/figure><\/div>\n\n\n\n

Ablation Study – DA, AW, CMC with bounding box embeddings<\/h2>\n\n\n\n
\"\"
Using simple <\/figcaption><\/figure><\/div>\n\n\n\n

Final Results <\/strong>– Deep OC-SORT<\/strong><\/h2>\n\n\n\n

Validation – MOT17 and 20<\/strong><\/h4>\n\n\n\n
\"\"<\/figure>\n\n\n\n
Test Set – Official MOT Challenge Leaderboard <\/h6>\n\n\n\n
\"\"<\/figure>\n\n\n\n

We need to perform additional benchmarking and submissions on DanceTrack, along with extra analysis of our methods and their failure modes, but our goal is to refine and publish our results after the end of our MSCV program. <\/p>\n","protected":false},"excerpt":{"rendered":"

Summary – We achieve SoTA in MOT20 and 2nd place in MOT17 on the official MOT Challenge Leaderboard Our failure analysis on SoTA algorithms such as OC-SORT led us to establish that occlusion-based false negatives and ID swaps are the biggest source of error for SORT based trackers. Additionally, we found that naively adding appearance … <\/p>\n

Continue reading “Results”<\/span><\/a><\/p>\n","protected":false},"author":142,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-19","page","type-page","status-publish","hentry"],"yoast_head":"\nResults - Unsupervised Online Human Detection and Tracking<\/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\/2022team14\/results\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Results - Unsupervised Online Human Detection and Tracking\" \/>\n<meta property=\"og:description\" content=\"Summary – We achieve SoTA in MOT20 and 2nd place in MOT17 on the official MOT Challenge Leaderboard Our failure analysis on SoTA algorithms such as OC-SORT led us to establish that occlusion-based false negatives and ID swaps are the biggest source of error for SORT based trackers. Additionally, we found that naively adding appearance … Continue reading "Results"\" \/>\n<meta property=\"og:url\" content=\"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/\" \/>\n<meta property=\"og:site_name\" content=\"Unsupervised Online Human Detection and Tracking\" \/>\n<meta property=\"article:modified_time\" content=\"2022-12-21T04:48:17+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-content\/uploads\/sites\/69\/2022\/12\/Screenshot-2022-12-20-110852-1024x306.png\" \/>\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\\\/2022team14\\\/results\\\/\",\"url\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/results\\\/\",\"name\":\"Results - Unsupervised Online Human Detection and Tracking\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/results\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/results\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/wp-content\\\/uploads\\\/sites\\\/69\\\/2022\\\/12\\\/Screenshot-2022-12-20-110852-1024x306.png\",\"datePublished\":\"2022-04-29T01:44:05+00:00\",\"dateModified\":\"2022-12-21T04:48:17+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/results\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/results\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/results\\\/#primaryimage\",\"url\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/wp-content\\\/uploads\\\/sites\\\/69\\\/2022\\\/12\\\/Screenshot-2022-12-20-110852.png\",\"contentUrl\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/wp-content\\\/uploads\\\/sites\\\/69\\\/2022\\\/12\\\/Screenshot-2022-12-20-110852.png\",\"width\":1466,\"height\":438},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/results\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Results\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/#website\",\"url\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/\",\"name\":\"Unsupervised Online Human Detection and Tracking\",\"description\":\"Students - Adnan Ahmad, Gerard Maggiolino | Advisors - Kris Kitani (CMU), Rahul Nallamothu (Meta), Sam Pepose (Meta)| Sponsor - Meta \",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/mscvprojects.ri.cmu.edu\\\/2022team14\\\/?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":"Results - Unsupervised Online Human Detection and Tracking","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\/2022team14\/results\/","og_locale":"en_US","og_type":"article","og_title":"Results - Unsupervised Online Human Detection and Tracking","og_description":"Summary – We achieve SoTA in MOT20 and 2nd place in MOT17 on the official MOT Challenge Leaderboard Our failure analysis on SoTA algorithms such as OC-SORT led us to establish that occlusion-based false negatives and ID swaps are the biggest source of error for SORT based trackers. Additionally, we found that naively adding appearance … Continue reading \"Results\"","og_url":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/","og_site_name":"Unsupervised Online Human Detection and Tracking","article_modified_time":"2022-12-21T04:48:17+00:00","og_image":[{"url":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-content\/uploads\/sites\/69\/2022\/12\/Screenshot-2022-12-20-110852-1024x306.png","type":"","width":"","height":""}],"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\/2022team14\/results\/","url":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/","name":"Results - Unsupervised Online Human Detection and Tracking","isPartOf":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/#website"},"primaryImageOfPage":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/#primaryimage"},"image":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/#primaryimage"},"thumbnailUrl":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-content\/uploads\/sites\/69\/2022\/12\/Screenshot-2022-12-20-110852-1024x306.png","datePublished":"2022-04-29T01:44:05+00:00","dateModified":"2022-12-21T04:48:17+00:00","breadcrumb":{"@id":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/#primaryimage","url":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-content\/uploads\/sites\/69\/2022\/12\/Screenshot-2022-12-20-110852.png","contentUrl":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-content\/uploads\/sites\/69\/2022\/12\/Screenshot-2022-12-20-110852.png","width":1466,"height":438},{"@type":"BreadcrumbList","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/results\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/"},{"@type":"ListItem","position":2,"name":"Results"}]},{"@type":"WebSite","@id":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/#website","url":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/","name":"Unsupervised Online Human Detection and Tracking","description":"Students - Adnan Ahmad, Gerard Maggiolino | Advisors - Kris Kitani (CMU), Rahul Nallamothu (Meta), Sam Pepose (Meta)| Sponsor - Meta ","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/?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\/2022team14\/wp-json\/wp\/v2\/pages\/19","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-json\/wp\/v2\/users\/142"}],"replies":[{"embeddable":true,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-json\/wp\/v2\/comments?post=19"}],"version-history":[{"count":4,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-json\/wp\/v2\/pages\/19\/revisions"}],"predecessor-version":[{"id":127,"href":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-json\/wp\/v2\/pages\/19\/revisions\/127"}],"wp:attachment":[{"href":"https:\/\/mscvprojects.ri.cmu.edu\/2022team14\/wp-json\/wp\/v2\/media?parent=19"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}