We were the first group to revisit the noise modeling problem and address it with machine learning. We’ve been approached by many companies Abdel has been approached by Samsung. Then you could train another model to denoise using this new knowledge, using this new model.
#Machine learning image cleaner how to#
MB: The beauty of this is, if you’re a company that wants to denoise images, you first need information on how to predict noise. Q: Many tech companies would be very interested in this. This is how the computer learned how to generate the noise. He would have, say, 500 noisy images and one clean image. Then he fed the information into the computer model and calculated an average. He set the camera on a tripod, very still in a special room, under special lighting, and took these images. He sat in the lab and captured 30,000 images. Could the computer generate the type of noise that a particular sensor would produce? We wanted to know if we could predict the type of noise that occurs on different sensors. MB: Broadly speaking, we’re trying to mathematically model noise. This is beneficial to computer vision because in computer vision, and deep learning, we need more data to make machine learning algorithms work. As a result, the computer could learn the characteristics of noise so it could generate more examples of noisy images. In our research, we collected lots of images, noisy images and their corresponding clean images, so the computer model could learn the mapping between the noisy images and the clean images. Q: How did the Noise Flow model work, and how could this benefit computer vision and deep learning?ĪA: The Noise Flow model is a type of machine learning – teaching computers how to solve a problem by providing many examples. We found that the mathematical models used to characterize noise in the literature were outdated – the models were for older, larger sensors, not for the sensors used in smartphones. Ten years ago, they didn’t have these tiny sensors. MB: Abdel’s method was in reverse he didn’t try to generate a clean image, he sought to understand the noisy image.ĪA: We studied older literature, papers from MIT and Google, and we saw that they had models for this, but the research hadn’t been updated in years – especially important with today’s smart cameras. But we took a different approach: we wanted to model, characterize or understand the noise in order to remove it more efficiently. Q: What was the objective of this research?ĪA: For computer vision or photography, the objective is to get rid of, or minimize, noise.
(Source: Smartphone Image Denoising Dataset (SIDD): ) The clean image to the right is heavily processed to minimize noise. The noisy image to the left is a common output of a smartphone camera in a dim environment. For small digital cameras, especially in dim environments, one of the trade-offs is a noisier image.
In this paper, we looked at noise from the perspective of smartphone cameras with very small sensors. For small digital cameras, one of the trade-offs is more inherent noise – a grainer or noisier image. The chance of having more noise is much higher. Now that camera sensors are smaller and smaller – on your phone, for example – so much less light gets in. MB: The camera sensors inherently have noise. If you’re in a low-light environment, you have more noise. You’re measuring the amount of light that gets in, in pixels. We all use digital cameras but, in some situations – when we’re in a dark room, for example – this negatively affects the quality of the image. Q: What is image noise and why is it problematic?ĪA: Image noise is an undesired effect that we see in digital images. This work was undertaken while Abdelhamed was at Borealis AI in an internship, supported by the Mitacs Accelerate Program.Ībdelhamed and Brown discuss this work and its impact with Brainstorm. This research was funded by VISTA and the Natural Sciences and Engineering Research Council of Canada (NSERC).
0 kommentar(er)
