17312276. AI DOWNSCALING APPARATUS AND OPERATING METHOD THEREOF, AND AI UPSCALING APPARATUS AND OPERATING METHOD THEREOF simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
AI DOWNSCALING APPARATUS AND OPERATING METHOD THEREOF, AND AI UPSCALING APPARATUS AND OPERATING METHOD THEREOF
Organization Name
Inventor(s)
Wookhyung Kim of Suwon-si (KR)
AI DOWNSCALING APPARATUS AND OPERATING METHOD THEREOF, AND AI UPSCALING APPARATUS AND OPERATING METHOD THEREOF - A simplified explanation of the abstract
This abstract first appeared for US patent application 17312276 titled 'AI DOWNSCALING APPARATUS AND OPERATING METHOD THEREOF, AND AI UPSCALING APPARATUS AND OPERATING METHOD THEREOF
Simplified Explanation
The patent application describes an AI-based system for upscaling low-resolution images to high-resolution images. The system uses two deep neural networks (DNNs) to achieve this.
- The first DNN downscales an original image to create a low-resolution version.
- The second DNN then upscales the low-resolution image to a high-resolution image.
- The second DNN is trained to minimize the difference between the original training image and the restored images obtained through downscaling, upscaling, and retranslating.
Potential applications of this technology:
- Enhancing the quality of low-resolution images in various industries such as photography, video production, and medical imaging.
- Improving the visual experience in virtual reality (VR) and augmented reality (AR) applications by upscaling lower quality images or videos.
Problems solved by this technology:
- Addressing the issue of low-resolution images by providing a method to upscale them to a higher resolution.
- Overcoming the limitations of traditional upscaling techniques that often result in blurry or pixelated images.
Benefits of this technology:
- Provides a more accurate and detailed representation of low-resolution images.
- Enhances the visual quality of images without the need for manual editing or intervention.
- Can be applied to a wide range of industries and applications, improving the overall user experience.
Original Abstract Submitted
An artificial intelligence (AI) upscaling apparatus for upscaling a low-resolution image to a high-resolution image includes: a memory storing one or more instructions; and a processor configured to execute the one or more instructions stored in the memory, wherein the processor is configured to: obtain a second image corresponding to a first image, which is downscaled from an original image by an AI downscaling apparatus by using a first deep neural network (DNN); and obtain a third image by upscaling the second image by using a second DNN corresponding to the first DNN, and wherein the second DNN is trained to minimize a difference between a first restored image, which results from applying no pixel movement to an original training image, and second restored images, which result from downscaling, upscaling, and subsequently retranslating one or more translation images obtained by applying pixel movement to the original training image.
