Welcome to the 【AI Daily】 section! This is your daily guide to exploring the world of artificial intelligence. Every day, we present you with the latest hot topics in the AI field, focusing on developers to help you understand technological trends and innovative AI product applications.
Fresh AI products Click to learn more: https://top.aibase.com/
1. Keling AI Platform 1.5 Model Upgrade: New Face Model Feature and Added "Standard Mode"
The 1.5 model upgrade of the Keling AI platform brings several innovative features, especially the introduction of the face model and standard mode, greatly enhancing the user creation experience.
【AiBase Highlights:】
🎨 Enhanced high-quality mode, allowing users to specify motion trajectories and add still areas for finer control.
⚡ New standard mode for quickly generating 720p videos, significantly reducing creation time and offering great value.
👤 Face model feature is available to members, solving the challenge of maintaining face IDs, supporting on-demand video generation.
2. Runway Introduces "Expand Video" Feature for Video Extension
Runway's "Expand Video" feature brings new possibilities for video creation, allowing users to extend video content through simple text prompts, creating more vivid and dynamic visual effects. This launch marks another advancement in video generation technology.
【AiBase Highlights:】
🎥 New feature launched: RunwayML's "Expand Video" allows users to extend content based on text prompts from the original video.
⚙️ Dynamic effects realization: Users can create dynamic effects such as quick zooming and panning, enhancing the visual experience of the video.
🚀 Increased generation speed: The new Gen-3Alpha model significantly improves generation speed, meeting users' rapid creation needs.
3. Huawei Mate 70 Series to Support AI Air Transfer Demonstrated by Yu Chengdong
The Huawei Mate 70 series introduces the AI air transfer feature, allowing users to easily transfer content with gestures, enhancing convenience of use.
【AiBase Highlights:】
🤖 AI air transfer: Users can transfer content from the Mate 70 phone to a tablet with simple gestures, making it easy to operate.
✋ Magic gesture operation: When hands are inconvenient, users can perform operations like page turning with gestures, increasing flexibility.
🌐 Native HarmonyOS: The Mate 70 series is the first to feature native HarmonyOS, marking progress in China's mobile operating system field.
4. Lightricks Launches Open Source Real-time Video Generation Model LTX-Video, Generating a Video in Just 4 Seconds!
The LTX Video model launched by Lightricks marks a new era in video generation technology with its ultra-efficient generation speed and outstanding performance.
【AiBase Highlights:】
🚀 The LTXV model is based on a diffusion transformer architecture with 2 billion parameters, generating videos faster than playback, producing a 5-second video in just 4 seconds.
💻 Utilizing bfloat16 precision, it runs efficiently on mainstream graphics cards, optimizing memory usage to ensure video quality.
🌍 LTXV seamlessly integrates with ComfyUI, lowering the usage barrier and democratizing creative technology, providing limitless possibilities for creators.
Details link: https://www.lightricks.com/ltxv
5. EchoMimicV2: Generate "Similar" Digital Humans by Inputting Images, Audio, and Gesture Videos
The emergence of EchoMimicV2 marks a new breakthrough in human animation generation technology. By simplifying input conditions, this technology not only enhances the detail of animations but also provides greater flexibility for creators. As someone passionate about animation technology, I am excited about this innovative approach and look forward to its potential applications in the future.
【AiBase Highlights:】
✨ EchoMimicV2 achieves high-quality half-body human animation generation by simplifying control conditions.
🎨 Adopting the Audio-Pose Dynamic Harmonization strategy, it enhances the detail and expressiveness of animations.
📊 New benchmark evaluation methods show that EchoMimicV2 outperforms existing technologies in effectiveness.
6. Audio-Driven Face Video Generation Technology JoyVASA
JoyVASA is an emerging audio-driven image animation technology aimed at improving video quality and lip-sync accuracy. Its innovative two-stage design makes it possible to generate longer animated videos while supporting both human and animal animations, showcasing excellent multilingual support capabilities.
【AiBase Highlights:】
🎨 JoyVASA technology enables the generation of longer animated videos by decoupling facial expressions from 3D models.
🔊 This technology can generate action sequences based on audio cues, possessing dual capabilities for human and animal animations.
🌐 JoyVASA is trained on Chinese and English datasets, providing multilingual support for users worldwide.
Details link: https://jdh-algo.github.io/JoyVASA/
7. PicMenu: Visualizing Menus with AI
PicMenu is an innovative AI menu visualization tool that transforms traditional paper menus into vivid and intuitive digital displays through simple photo-taking operations. It provides a new solution for the digital transformation of the restaurant industry, demonstrating strong practicality and technical prowess.
【AiBase Highlights:】
📸 PicMenu can automatically recognize and extract dish information through photos, generating high-quality digital menu displays.
🌍 This tool employs multiple AI technology stacks to ensure the efficiency and accuracy of dish information extraction and image generation.
🔧 The project is open-source, allowing developers to access the source code via GitHub for personalized customization and deployment.
Details link: https://www.picmenu.co/
8. NVIDIA Releases Edify 3D Generation Model: Generate High-Quality 3D Assets in 2 Minutes
NVIDIA's latest Edify3D technology has made significant breakthroughs in 3D asset generation, capable of producing high-quality 3D models in a short time, greatly enhancing creative efficiency.
【AiBase Highlights:】
🎨 Edify3D technology can generate high-quality 3D models in two minutes, supporting complete UV mapping and 4K textures, suitable for various industries.
🔍 This technology combines multi-view diffusion models and Transformer reconstruction techniques to ensure generated models have precise mesh structures and high-resolution textures.
🌐 Edify3D also supports generating complex 3D scenes based on text prompts, providing robust artistic design and 3D modeling support.
Details link: https://research.nvidia.com/labs/dir/edify-3d/
9. Smarter! Xiaomi Smart Home Screen Series Products Push Super Large Model for Xiao Ai
Xiaomi has upgraded the large model functionality of its smart home screen series products, introducing new features such as super Q&A and casual chat, enhancing user experience.
【AiBase Highlights:】
🔧 The upgrade covers six devices, including Xiaomi Smart Home Screen 10 and Xiaomi Xiao Ai Touchscreen Speaker Pro 8.
📱 Users need to upgrade their speakers to the latest version and set the large model feature in the app.
🌐 The new feature supports text creation and real-time search, allowing users to ask various questions directly to the speaker.
10. REDMI K80 Pro Equipped with Xiaomi AISP 2.0 Imaging Algorithm: Integrating Four Major Models
Xiaomi today disclosed the imaging configuration details of the REDMI K80 series smartphones, marking a comprehensive enhancement in the photography capabilities of this series.
【AiBase Highlights:】
📷 The REDMI K80 series features a flagship high-dynamic main camera and a floating telephoto lens, meeting high-quality photography needs.
⚙️ The K80 Pro is equipped with Xiaomi AISP 2.0 technology, supporting 150 full-algorithm high-quality quick shots, enhancing the shooting experience.
💰 The standard version and Pro version maintain consistency in the main camera and experience but do not include the high-cost floating telephoto lens, providing a more cost-effective option.
11. Pang Donglai Issues Statement: Warning Against Unauthorized Use of AI to Generate Yu Donglai's Voice
Pang Donglai Commerce Group has issued a warning against unauthorized use of its founder's voice, emphasizing that legal measures will be taken to protect its rights.
【AiBase Highlights:】
🔊 Multiple accounts are using AI to generate voices similar to Yu Donglai for illegal editing and tampering.
⚖️ Pang Donglai emphasizes that this behavior infringes copyright and creates false information, disrupting the online ecosystem.
🚫 The company demands that relevant accounts immediately cease infringement and states that legal actions will be taken.
12. Global Generative AI Chip Market Expected to Reach $86.24 Billion by 2028
According to the latest report, the generative artificial intelligence chip market is experiencing rapid growth, with an expected market size of $86.24 billion by 2028. This growth reflects a surge in demand for computing power and the popularity of cloud computing.
【AiBase Highlights:】
🌟 The market size is expected to reach $86.24 billion by 2028, with an average annual growth rate of 27.9%.
💻 Major driving factors include increased demand for computing power, the development of deep learning, and the popularity of cloud computing.
📈 Qualcomm has launched a new generation of chips to enhance generative AI capabilities in smartphones, driving market innovation.