--- title: Computer vision models no longer need labels. | daily.dev description: Self-supervised vision models like DINOv2 can learn high-quality visual representations without any human-generated labels. By comparing embedding vectors from... canonical: https://daily.dev/posts/computer-vision-models-no-longer-need-labels--e8gruzovn twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Computer vision models no longer need labels. | daily.dev og:description: Self-supervised vision models like DINOv2 can learn high-quality visual representations without any human-generated labels. By comparing embedding vectors from... og:url: https://daily.dev/posts/computer-vision-models-no-longer-need-labels--e8gruzovn og:image: https://api.daily.dev/og/posts/e8GRUZOVN.png og:image:alt: Computer vision models no longer need labels. og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Computer vision models no longer need labels. **[Welch Labs](https://daily.dev/sources/welchlabsvideo)** · 1 min read · 0 upvotes · 0 comments ## Summary Self-supervised vision models like DINOv2 can learn high-quality visual representations without any human-generated labels. By comparing embedding vectors from image patches, DINOv2 can effectively segment objects (hands, balls, cats, books) from backgrounds purely through unsupervised learning, demonstrating remarkable representation quality. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [Do Labels Make AI Blind? Self-Supervision Solves the Age-Old Binding Problem](https://daily.dev/posts/do-labels-make-ai-blind-self-supervision-solves-the-age-old-binding-problem-qiwivfcrm) · Towards Data Science · 0 upvotes · 0 comments - [Grounding DINO: Open Vocabulary Object Detection on Videos](https://daily.dev/posts/grounding-dino-open-vocabulary-object-detection-on-videos-g6hotulhw) · PyImageSearch · 0 upvotes · 0 comments - [Expanding visual search for online art collections](https://daily.dev/posts/expanding-visual-search-for-online-art-collections-1qazg90yx) · Q42 Engineering · 0 upvotes · 0 comments - [Optimizing Semiconductor Defect Classification with Generative AI and Vision Foundation Models](https://daily.dev/posts/optimizing-semiconductor-defect-classification-with-generative-ai-and-vision-foundation-models-knmw1j4r8) · NVIDIA Developer · 0 upvotes · 0 comments --- Tags: [#machine-learning](https://daily.dev/tags/machine-learning), [#computer-vision](https://daily.dev/tags/computer-vision), [#embeddings](https://daily.dev/tags/embeddings) [View this post on daily.dev](https://daily.dev/posts/computer-vision-models-no-longer-need-labels--e8gruzovn) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Computer vision models no longer need labels.","url":"https://daily.dev/posts/computer-vision-models-no-longer-need-labels--e8gruzovn","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/computer-vision-models-no-longer-need-labels--e8gruzovn"},"datePublished":"2026-07-08T00:23:51.274Z","dateModified":"2026-09-14T07:11:56.771Z","description":"Self-supervised vision models like DINOv2 can learn high-quality visual representations without any human-generated labels. By comparing embedding vectors from...","image":"https://i.ytimg.com/vi/MHqeadiJ590/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/MHqeadiJ590/sddefault.jpg","isAccessibleForFree":true,"articleSection":"Welch Labs","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Welch Labs","logo":"https://media.daily.dev/image/upload/s--FpOVzn-9--/f_auto/v1747295382/logos/welchlabsvideo","url":"https://daily.dev/sources/welchlabsvideo"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/computer-vision-models-no-longer-need-labels--e8gruzovn","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"machine-learning,computer-vision,embeddings","timeRequired":"PT1M","video":{"@type":"VideoObject","name":"Computer vision models no longer need labels.","description":"Self-supervised vision models like DINOv2 can learn high-quality visual representations without any human-generated labels. By comparing embedding vectors from...","thumbnailUrl":"https://i.ytimg.com/vi/MHqeadiJ590/sddefault.jpg","uploadDate":"2026-07-08T00:23:51.274Z","duration":"PT1M","url":"https://api.daily.dev/r/e8GRUZOVN","embedUrl":"https://www.youtube.com/embed/MHqeadiJ590"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Welch Labs","item":"https://daily.dev/sources/welchlabsvideo"},{"@type":"ListItem","position":3,"name":"Computer vision models no longer need labels."}]} ```