IMAGE DEBLURRING VIA SELF-SUPERVISED MACHINE LEARNING: abstract simplified (17655427)
The abstract describes a technique for using machine learning to deblur images. The process involves receiving an input image and using a neural network to generate a deblurred image. This is done by first processing the input image to create a feature tensor using a portion of the neural network. Then, a motion mask is generated by processing the feature tensor using another portion of the neural network. Finally, the deblurred image is generated by processing the feature tensor and the motion mask using yet another portion of the neural network.