Twelve Labs

Video AI just got 10x faster: By harnessing Oracle Cloud Infrastructure, Twelve Labs is redefining how machines understand video content! We specialize in training large multimodal AI systems that understand video content, just like humans do. 🎥 Our models learn by analyzing countless hours of video, enabling developers and enterprises to pinpoint precise moments within vast archives or summarize entire collections efficiently. With Oracle's high-performance RDMA network and advanced GPU configurations, we've achieved training efficiencies that are 5 to 10 times faster, dramatically accelerating our time to market. 🌐 Through our expanding compute capabilities with Oracle Cloud, we're committed to delivering best-in-class multimodal AI models for video understanding to tens of thousands of developers worldwide. https://lnkd.in/gCpiWPhJ #VideoAI

Key highlights from SMPTE Media Technology Summit! Our own Simran Butalia spoke to a room full of filmmakers, technologists, and builders at Ovation Hollywood. Simran explored how Twelve Labs models are accelerating content production by efficiently managing and analyzing large volumes of video. Our models enhance productivity and empower creators by significantly accelerating the process of organizing and finding video content, from dailies to full game footage and behind-the-scenes material. Thank you to everyone who joined us and gained insights into how AI is helping media production! 🙌 #SMPTE2024 #MediaTech #VideoAI

In the 63rd session of #MultimodalWeekly, we have three exciting presentations on video summarization, compositional video understanding, and tracking everything. ✅ Son Jaewon from Sungkyunkwan University will discuss CSTA. This method stacks each feature of frames from a single video to form image-like frame representations and applies 2D CNN to these frame features: https://lnkd.in/gH2f3F76 ✅ Jinwoo Ahn from Hanyang University will discuss his paper Compositional Video Understanding. The proposed method includes a new transformer architecture capable of learning spatiotemporal graphs, and a compositional learning method to learn disentangled features for each semantic unit: https://lnkd.in/gpjU5TaE ✅ Siyuan Li from ETH Zürich will discuss his work MASA, a novel method for robust instance association learning, capable of matching any objects within videos across diverse domains without tracking labels. Leveraging the rich object segmentation from the Segment Anything Model (SAM), MASA learns instance-level correspondence through exhaustive data transformations: https://lnkd.in/dqN_zdfn Register for the webinar here: https://lnkd.in/gJGtscSH 👈

Continuing to Innovate! Twelve Labs and Mimir are enhancing the way users engage with video content through our powerful integration. This collaboration brings our advanced semantic search capabilities right into the Mimir interface, allowing for intuitive searches based on context, meaning, and more—all in natural language. 🎥 By leveraging Twelve Labs' state-of-the-art video AI with Mimir’s comprehensive asset management platform, we’re enabling organizations across various sectors to effectively navigate and utilize their extensive video archives. 🔗 Discover more about how our collaboration with Mimir is transforming video asset management and enhancing content discoverability. https://lnkd.in/gC26K6yP #VideoAI #TechInnovation #MediaTech

Imagine a world where every video becomes an interactive quiz, effortlessly transforming passive viewing into active learning 🎓 Our new tutorial by Hrishikesh Yadav explores the Video Content Quiz Generator, an application designed to revolutionize the learning experience associated with video content. Powered by Twelve Labs APIs, this video-based tool automatically generates Multiple Choice Questions (MCQs) from video content, enhancing both learning and assessment. 👨🏫 Whether you're an educator seeking quick assessments or a content creator aiming to boost engagement, the Video Content MCQ Generator is your ally in fostering active learning. 🤝 ☑ Read the tutorial: https://lnkd.in/ggVRQua5 ☑ Watch the demo video: https://lnkd.in/gHAsDZ_t ☑ Explore the demo app: https://lnkd.in/guJEBUjH ☑ Play with it via this Replit template: https://lnkd.in/gSWUCY4E

~ New Webinar ~ The recording of #MultimodalWeekly 58 is up! Watch here: https://lnkd.in/gx2egvEY 📺 They discussed: 1️⃣ Enhancing Video Super-Resolution via Implicit Resampling-based Alignment (Kai Xu - National University of Singapore) 2️⃣ Benchmarking Multimodal LLMs (Bohao Li - The Chinese University of Hong Kong) Enjoy!

Excited to share our collaboration featured in the latest Milvus tutorial! Discover how Twelve Labs' Embed API, combined with Milvus' powerful vector database, transforms semantic video search. You can now search across multiple modalities, including text, audio, and video, and pinpoint exact moments from long videos. ✨ Dive into our comprehensive guide covering everything from setting up your environment to executing advanced features like hybrid and temporal video searches. https://lnkd.in/gZ-uqp4a #VideoAI

Twelve Labs will be at the SMPTE Media Technology Summit from October 21st to 24th at Ovation Hollywood, California! Join us at one of the premier non-commercial conferences in media technology. 🌟 We're looking forward to connecting with industry leaders and showcasing our latest innovations at booth 105. Whether you're a C-Suite executive or a media tech enthusiast, this event offers incredible opportunities to network, learn, and explore the future of media technologies. 👋 Stop by our booth to discover how Twelve Labs is driving advancements in media AI. https://lnkd.in/g5RZQt4w

What an electrifying weekend at LA's largest gen AI hackathon, Cerebral Beach Hacks! It was inspiring to see 500 hackers passionately tackle multimodal challenges inspired by real-world problems that our customers face. 🤝 A huge thank you to our partners AI LA, Kindo, Google, LlamaIndex, and others who joined forces to make this event a success. It was inspiring to watch brilliant minds use the Twelve Labs models to innovate and create solutions that push the boundaries of what AI can do. 🌟 Everyone brought their best to the table, equipped with access to our cutting-edge models, proving that together, we can accelerate progress and transform industries. #CerebralBeachHacks

Reflecting on a fantastic time at GopherCon Korea 2024! Twelve Labs was thrilled to connect with the vibrant community of Go developers in South Korea. It was a great opportunity to share insights with the talented Korean developer community and engage in meaningful discussions about the future of Go programming. 👩💻🧑💻 Our booth was buzzing with energy as we hosted office hours and demos, opening doors to exciting recruitment conversations and sharing insights into our pioneering video understanding technologies. 💬 We loved every minute of exchanging ideas and discussing potential career opportunities with the talented developers who stopped by. The passion and innovation in the room were truly inspiring! Thank you to everyone who joined us and made this event a memorable one. We're already looking forward to more interactions and breakthroughs at the next GopherCon! #GopherConKorea2024 #VideoAI