MediaTekandquick workerYesterday, we jointly announced the launch of the high-efficiency end-sideVideo Generation Technology.
This technology is a continuation and improvement of the video generation technology that debuted at the 2024 World Mobile Communications Conference. It combines Kuaishou’s AI model I2V (image to video) Adapter withMediaTek Dimensity 9300 and Dimensity 8300 mobile platformsAI computing power,Generate dynamic video from static images on the client side.
With this technology, users only need to select an image on their device and the app canIdentify people and scenes in photos, generate natural and smooth videos, and improve the creative expression and efficiency of video production.
I2V-Adapter is a lightweight adapter module proposed by the Kuaishou AI team that is suitable for the Stable Diffusion-based Vincent video diffusion model and is specifically used for image-to-video generation tasks.
According to reports, MediaTek Dimensity 9300 and Dimensity 8300 mobile platforms integrate NPU (neural network processing unit) designed for generative AI. NPU can accelerate the calculation of AI models. Combined with the adaptation and optimization of Kuaishou I2V Adapter model, it can achieve faster video generation speed while also having low power consumption characteristics.
As previously reported by IT Home, at the Dimensity Developer Conference held in May, MediaTek, together with partners such as Alibaba Cloud, Baichuan Intelligence, Transsion, Zero One Things, OPPO, Honor, vivo, and Xiaomi, jointly launched the "Dimensity AI Pioneer Plan" to help developers build generative AI on terminal devices equipped with Dimensity chips.