Η Chinese company AI Lab of Xiaomi he published a new document that describes an AI network called "DeepExposure", which enhances low quality images through machine learning. «Comparing with other methods", The researchers explain in their article,"our algorithm can restore most details in the original images while enhancing brightness and colors».
Technology Deep Exposure uses challenging networks and asynchronous learning to split low-quality images into sections called sub-images. The system calculates both the local and total reports for these sub-images, and then evaluates their quality before integrating them with the original image.
The end result is an image with improved exposure and detail, opening the door to future solutions that can automatically enhance low quality photos. It is possible that Xiaomi will one day offer this technology to its smartphones, which already offer other AI capabilities.
The researchers point out in their journal: “Due to the demand for good photos by ordinary users, the improvement of photo quality is beyond the capabilities of non-professional users, thus leading to the new trend of automatic techniques for retouching the image.".
The study shows that Deep Exposure could also be adjusted in the future to improve the tone and contrast of the image.
Η Xiaomi is not the only company that uses artificial intelligence to automatically enhance images. Earlier this year, her researchers Intel and the University of Illinois Urbana-Champaign analyze a deep neural network , capable of illuminating low light images without reducing their quality. Similarly, a technology called Deep Image Prior was revealed last year with the ability to recreate damaged parts of an image based on existing image elements.
[the_ad_group id = ”966 ″]
