Anjali Rauniyar’s Post

I had a quiet moment on Sunday (thanks to my kids deciding to sleep in) where I realized that every day I work with some form of AI in my role at Wisdom Analytics Inc. It's been incredibly transformative for me, including seeing our #AI pipeline plowing through massive amounts of data, creating these amazing models that answer questions that are out of bounds for others (shameless plug, I know!) #AI alone isn't going to change the world. It's the science in the Art+Science equation that has been around forever - it's just a different sort of science - and it's constantly being updated. In the video below, the discussion is around the use of OpenAI #sora - an AI video generation implementation - combined with the Unreal Engine - which I think has been one of the most influential game engines for many years. Game engines like Unreal have been used in the past for simulations in business, such as medical trials. With the overlay of Sora, there's significant potential for businesses to use this type of combination to create so many other applications. Think marketing, product showcases, virtual events... the list goes on. The video is a little "techy", but hopefully you will also see the potential AI has in changing how many aspects of life can operate. Let me know your thoughts! #wisdomanalytics #videoai https://lnkd.in/gwc9ZhPY

🌌 Unveiling the Future: OpenAI's #Sora Revolutionizes #AI with Unmatched Physics & Emergent Abilities! 🚀 Dive into the future with OpenAI's groundbreaking Sora model! 🌠 Wes Roth takes us on an exhilarating journey, revealing how Sora is not just a leap but a quantum jump in AI video generation. Imagine a world where AI-created videos are indistinguishable from reality. That's Sora for you! 🎥✨ Sora stands out by simulating 3D environments with physics so real, you'll have to do a double-take. From the delicate balance of ice cubes in a glass to the dynamic battles between pirate ships in a cup of coffee, Sora's capabilities are nothing short of magical. 🏴☠️☕️ But here's the kicker: what we see is just the tip of the iceberg. Behind the scenes, Sora is a powerhouse, blending data-driven physics engines with intuitive physics, long-horizon reasoning, and semantic grounding. It's not just creating videos; it's simulating entire worlds, real or fantastical. 🌍🔮 Dr. Jim Fan from #Nvidia hints at the complexity and potential of Sora, suggesting it might even leverage Unreal Engine 5 for its breathtaking simulations. This isn't just about video generation; it's about building the foundation for #AGI through intricate rendering and intuitive physics. 🧠💡 Join Wes Roth as he unpacks the marvel that is Sora, offering a glimpse into the future of AI that OpenAI is shaping. This isn't just an advancement; it's a revolution. And we're here for it. 🌟🛠️ 👍 30 min.⏱here: https://lnkd.in/eq3yCcgr #OpenAI #SoraRevolution #AIFuture #PhysicsSimulation #AGIPotential

#GenAI Video Accelerates in 2024 via Sora by OpenAI... Creating instant video from text... what are your views? This is all synthetic. Sora's a data-driven physics engine, a simulation of many worlds, real or fantastical. The simulator learns intricate rendering, "intuitive" physics, long-horizon reasoning + semantic grounding, by some denoising and gradient maths. This is no toy! #AI #DataDriven #PhysicsEngine #Simulation

Computer vision models are very good at seeing what you trained them to see. But, they are very bad at when you present an alien class (or a bad image of a known class), that was never shown to the model. It will hallucinate, similar to what LLMs models are doing now. It may for instance see a boat when you have a snake on water, technically boats are on water. They may see rabbits on grass with no rabbits, technically rabbits are seen on grass-like environment. #computervision #machinelearning #artificialintelliegence https://lnkd.in/dAf8KqUk

This week I've been spinning up our MVP. We've got some pretty serious commitments for our pilot programme, so I need to get to work...fast Our value prop depends on a lower cost to serve, faster agent response times and a great user experience. For this reason, I've been exploring AI and the tools that can help me move fast. Some of the best I've found are: 1. LangChain has let me build an AI workflow on top of open AIs general models. David Fava Showed me how you can use this to learn what your model is doing under the hood! 2. OpenAI Assistants API w Custom functions is mind blowing. In particular how easy it is to get a chat model to capture inputs from the user and pass it to my app. 3. OpenPipe Made it super easy and cheap to deploy a fine tuned model feat automatic deployment. Thanks for the tip Fred Johnson 4. I've used this model from the user vikhyatk for image to text. It's cheaper than the general LLMs and has better results for my use case. (https://lnkd.in/eQ4cf53S) In future I plan to spend more time being selective about our LLM infra. Have a good day!

OpenAI has introduced Swarm, an experimental framework intended to develop systems where multitudes of AI agents can jointly solve many complex problems much better. Unlike traditional single-agent models, Swarm takes center stage in orchestrating interactions among agents that allow them to scale efficiently by specializing. From big data analysis to complex simulations, Swarm offers scalability, flexibility, and parallel processing power for extending the state of the art in multiagent systems. See the full insights here: https://lnkd.in/gKWr6NRF

Imaging generating video just by writing prompts for them ! OpenAI just dropped a bombshell in the AI world with SORA - a groundbreaking leap in video generation models! 🚀 This game-changing release trains massive generative models on diverse video and image data, paving the way for a new era of AI-driven simulation. Check out this read : https://lnkd.in/d75DS_-E #OpenAI #SORA #VideoGeneration #AIInnovation 🤖

Video Metadata utilizes deep learning algorithm to detect and capture images of targets (human, vehicle and non-motor vehicle). It selects the images with the best quality and extract relevant attributes. Reports on targets' quantity can be exported by day, week, month and year. #ibs #dahua #distributor #vad #cctv #physicalsecurity #WizMind #VideoMetadata #videowall #nvr #vms #fullcolor #metadata

Mixing Houdini's proceduralism to generate synthetic data and train models with the power of running on Conductor on CoreWeave, kickstarts what VFX artists can do with AI/ML. DM me if you want to chat more about this. #sidefx #houdini #ai #ml #conductor #coreweave #vfx

In continuation to my earlier post on General World Models (GWMs), the research work titled as 𝗪𝗼𝗿𝗹𝗱𝗗𝗿𝗲𝗮𝗺𝗲𝗿 represents a breakthrough in how AI systems learn to understand the visual world. By treating 🎖🎖🎖🎖🎖 WＯＲＬＤ　ＷＯＤＥＬＩＮＧ　ＬＩＫＥ　Ａ　ＬＡＮＧＵＡＧＥ　ＵＮＤＥＲＳＴＡＮＤＩＮＧ 🎖🎖🎖 problem, it can capture the complex ways that objects move and interact in diverse environments. 𝗣𝗿𝗼𝗯𝗹𝗲𝗺: Current world models, which help AI understand the dynamics of the world, are often too specialized for gaming, driving, or robotics. This limits their ability to understand the complex physics and motions of the real world. 𝗦𝗼𝗹𝘂𝘁𝗶𝗼𝗻: The proposed work in WorldDreamer presents a new type of world model designed to learn about general visual environments. 𝗞𝗲𝘆 𝗶𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻𝘀: 🏅 Treats world modeling as a visual sequence prediction problem 🏅 Leverages a Spatial Temporal Patchwise Transformer for efficient learning 🏅 Enables diverse tasks like text-to-video, image-to-video, and action-to-video generation This opens up exciting possibilities for more advanced video generation and editing tools. #AI #VideoGeneration #WorldModeling #GenAI #SORA #OpenAI #GWMs #GeneralWorldModels