 Hi there, as you may know, we have just released a new MPU, the STM32 MP2, which has some great feature for vision AI applications. This includes a versatile and compact ISP, an image signal processor, which is aimed at producing mind-blowing image quality, and a powerful neural processing unit to afloat CPU from AI model execution. Pour l'ISP, we are introducing an embedded world, a software tuning tool that will enable you to configure it and extract the finest image quality from any sensor. This is geared as saving development time and simplifying your system design as much as possible. We are calling it the STM32 ISP IQ Tune. It provides an interactive step-by-step guide through various ISP settings. It features a preview window that allows you to immediately witness the impact of your adjustment. At every stage, we are pre-set configurations and helpful advice to ensure you extract great image quality from any sensor. Check out our demo here. We are modifying the lighting conditions and the scenes illuminants within the light box to showcase how our auto white balance and auto exposure seamlessly adapt to changing contexts. Traditionally, customers would have to invest in services from image quality companies for ISP tuning. But now, with the ISP IQ Tune, we are proud to offer a unique solution on the market that is free of charge, with no code and user-friendly to tackle this daunting task. So, once you have got the best image quality, the next step is to ensure that you have the right neural network to complete your Vision AI use case. With the STM32 MP2, for example, you can leverage its 1.3-tops NPU to create advanced AI applications. And that's why we have made all our microprocessors available on STM32 Cubed AI Developer Cloud. The Developer Cloud is a set of its optimization tools and services to help you implement AI quickly and easily on your STM32. Building on its success for the STM32 microcontrollers introduced last year, we are confident that it will also boost AI development for our MPU customers. To get started, all you have to do is to connect and log in. A key feature of its benchmarking service using STM32 Board Farms World Ride. Within minutes, you can remotely obtain metrics on accuracy, latency, ram et flash size, toujours sans besoin pour d'autres hardware sur votre desk. The Developer Cloud s'occupe d'un STM32 Model Zoo sur GitHub. Il inclure une repositerie de modèles, un script de entraînement et un exemple d'application pour rendre la vie d'un data scientist plus facile. Si vous voulez savoir plus sur ces visions fantastiques et de développement de l'AI, please contact your local ST representative or check out st.com. Thanks for watching.