Generative AI (AI) is an AI technology that can generate new content based on existing data. It can create new content such as text, images, audio, video, and computer code by learning patterns and regularities in large datasets, and centers on the use of machine learning models, especially deep learning models, that are pre-trained with large amounts of unlabeled or labeled data in order to learn about the distribution and structure of the data. These models are called Foundation Models (FMs), and they are capable of performing a wide range of tasks such as image recognition, natural language processing, and so on. Large Language Models are a type of Foundation Models that specialize in language-related tasks such as text generation, classification, and conversation.
Generative AI models can be categorized into various types, including text-to-text, text-to-image, image-to-text, etc., and they have a wide range of applications in different domains. For example, text-to-text models can be used for machine translation, text summarization, dialog production, and story creation; text-to-image models can generate corresponding images based on text descriptions, which are applied to image generation, annotation, and editing. In addition, generative AI has significant application potential in industries such as healthcare, automotive manufacturing, and entertainment media. The importance of generative AI lies in its ability to drive innovation, increase efficiency, and improve user experience.
The rise of generative AI is impacting the future of industries, with edge computing becoming the key to enabling AI. At the upcoming Embedded World 2024, Coretronic will showcase several computing platforms based on Qualcomm's processors and self-developed AI SOM solutions. These solutions are designed to help customers realize a variety of smart terminal applications, including robots, AI Boxes, industrial PCs, and other AI-enabled IoT products. Qualcomm's RB3 Gen 2 platform, based on the Qualcomm QCS6490 processor, provides powerful AI processing capabilities and computer vision functions, while supporting multiple operating systems to meet the diverse operating system needs of smart terminals. In addition, Coretronic will also exhibit its UAV SOM solution, which uses Qualcomm QRB5165 processor with high performance, miniaturization, and energy efficiency, and supports long-distance communication module with communication distance up to 10km or more, making it the world's first commercial UAV hardware solution to integrate Qualcomm RB5/QCN9074. This will help customers focus on the development of flight control and related application software, shorten product development time, and significantly improve the efficiency of AI computing and the quality of long-distance mapping, which is suitable for intelligent surveillance, search and rescue, and other fields.
Dr. Kuei-Ting Chen, Chief Technology Officer of CIIC, said that the cooperation with Qualcomm Technologies is not only a partnership, but also represents the integration of the strengths of both companies. By combining Qualcomm's wireless communications, generative AI and SOM technologies, it demonstrates both companies' dedication to driving innovation and their commitment to providing cutting-edge solutions to meet customers' evolving needs.
Dev Singh, vice president of business development and head of the robotics and industrial automation business at Qualcomm Technologies, also recognized the partnership and looks forward to demonstrating the power of edge computing and generative AI at Embedded World 2024, bringing customers advanced computing platforms and AI solutions that unleash the full potential of smart devices in robotics and AIoT smart devices in applications. of its full potential.
