AZHAR的动态

This picture of me being freezing at the Union Square Holiday Market was taken with a Pixel 9, using the long exposure feature. It leverages AI-driven #computationalimaging to achieve this effect. The camera takes multiple frames in very quick succession and then uses #AI to separate which elements are static and which ones are moving. It uses AI to determine the direction and speed of the motion and then applies motion blur to the moving objects. It uses AI to stabilize the static elements across frames so you can use the feature handheld. It also uses AI to optimize the image for things like brightness, contrast and sharpness. I think people who purport not to like AI maybe don't realize most of us are using it every day. #googlepixel

Meta just released Llama 3.2, a major update to its AI model lineup! This new version includes both lightweight models designed for mobile and edge devices, and larger models with vision capabilities. For the first time, Llama models can now analyze images along with text, opening up more possibilities for AI applications. The smaller models (1B and 3B) are optimized for on-device tasks, while the larger 11B and 90B models can perform advanced tasks like image recognition and visual question answering. Meta is making these models available for free, pushing open-source AI even further! #AI #Llama3 #MetaAI #TechInnovation #OpenSourceAI #MultimodalAI

Google Shatters AI Boundaries: Veo 2 & Imagen 3 Dominate Generative Media Google has unveiled groundbreaking AI models: Veo 2, a revolutionary video generation model, and Imagen 3, an image model that sets new standards for quality and realism. The AI landscape is rapidly evolving, and Google continues to lead the charge with its latest advancements. Veo 2 and Imagen 3 demonstrate a significant leap forward in generative media, pushing the boundaries of what's possible with AI. Key Points: ??Veo 2: Generates high-resolution 4K videos (720p at launch). Excels in cinematic control and physics simulation, resulting in more realistic and believable outputs. Significantly reduces hallucinations, leading to more accurate and predictable results. Outperforms competitors like OpenAI's Sora in human evaluations. ??Imagen 3: Delivers unparalleled color vibrancy, composition, and detail across various artistic styles. Accurately interprets and fulfills complex user prompts with exceptional precision. Outperforms Midjourney, Flux, and Ideogram in human evaluations for preference, visual quality, and prompt adherence. Why This is a Game Changer: Veo 2 and Imagen 3 have the potential to revolutionize how we create and consume media. From filmmaking and advertising to art and education, these models open up new avenues for creativity and innovation. https://lnkd.in/evV_CHXc #googleai #veo2 #imagen3 #ai #generativeai #videogeneration #imageGeneration #innovation #deeplearning #future

This looks like an interesting use of Ai

Project Astra represents a significant leap forward in the quest to create Artificial Intelligence. It has the potential to revolutionize how we interact with technology in our daily lives!!!. However, with such advancements also come's ethical considerations and questions about privacy and data usage. As we move forward, it's crucial to ensure that these powerful AI technologies are developed and deployed responsibly, with careful consideration given to their societal impacts!!!! #AI #google #Technology #Future

So what do you think of this latest AI assistant?

Google introduced Project Astra! Project Astra is Google DeepMind's vision for a future AI assistant that can see, hear, and understand the world around us in real time (similar to a human). OpenAI got a real competitor here! The agent takes in a constant stream of audio and video input. It can reason about its environment in real time and interact with the user in a conversation about what it is seeing. There are two continuous takes: one with the prototype running on a Google Pixel phone and another on a prototype glasses device. The future is here. This is just the beginning. Follow Endrit Restelica to stay up to date with AI. #ai #googleio

Google?introduced Project Astra! Project Astra is?Google DeepMind's vision for a future AI assistant that can see, hear, and understand the world around us in real time (similar to a human). OpenAI?got a real competitor here! The agent takes in a constant stream of audio and video input. It can reason about its environment in real time and interact with the user in a conversation about what it is seeing. There are two continuous takes: one with the prototype running on a Google Pixel phone and another on a prototype glasses device. The future is here. This is just the beginning. Follow?Endrit Restelica?to stay up to date with AI. #ai?#googleio

The last 48 hours in AI have been intense! Google I/O kicked off yesterday with a wide array of AI announcements, headlined by Project Astra. This new AI assistant promises to change how we interact with technology significantly. Here are some key takeaways from Day 1 at I/O: ? Project Astra: A real-time, multimodal AI assistant that can see, understand, and assist with various tasks. ? Gemini 1.5 Flash: A faster model for tasks like summarization and captioning. ? Veo: Generates video from text prompts. ?Gemini Nano: Improved performance for mobile devices. ? Gemini Pro: Doubles the context window to 2 million tokens. ? Gemini Live: A voice-only assistant for more natural conversations. ? New Lens Features: Enables web searches via narrated video. Check out Techcrunch’s video summary of Day 1 below! #AI #GoogleIO #ProjectAstra?