Abstract: Techniques are disclosed for training and applying a denoising model. The denoising model includes multiple specialized denoisers and a generalizer, each of which is a machine learning model. The specialized denoisers are trained to denoise images associated with specific ranges of noise parameters. The generalizer is trained to generate per-pixel denoising kernels for denoising images associated with arbitrary noise parameters using outputs of the specialized denoisers. Subsequent to training, a noisy image, such as a live-action image or a rendered image, can be denoised by inputting the noisy image into the specialized denoisers to obtain intermediate denoised images that are then input, along with the noisy image, into the generalizer to obtain per-pixel denoising kernels, which can be normalized and applied to denoise the noisy image.

Abstract: A method, and optical network management system is provided for monitoring an optical transmission system, including detecting degradation of an electrical signal converted from an optical signal transported via a plurality of optical network elements within the optical transmission system. The method, and optical network management system additionally perform the step of correlating the degradation in the electrical domain with degradation of the optical transmission system in an optical domain based a plurality of threshold values associated with an optical performance parameter corresponding each of the optical network elements. The method, and optical network management system further perform the step of determining whether a cause of the degradation in the electrical domain stems from one or more of the optical network elements according to the correlation.

Abstract: A method, and optical network management system is provided for monitoring an optical transmission system, including assigning a plurality of threshold values corresponding to a performance parameter associated with the optical transmission system. The method, and optical network management system further perform the step of identifying degradation of the optical transmission system based on which of the threshold values being exceeded.

Abstract: A method, and optical network management system is provided for monitoring an optical transmission system, including detecting an error condition in an electrical domain of the optical transmission system, and collecting data associated with optical performance of one or more optical network elements in response to the detected electrical degradation. The collected data is used to determine whether the one or more of the optical network elements are a source of the electrical degradation.

Abstract: A method, and optical network management system, including detecting, in an electrical domain, a change in performance of an optical transmission system, and determining a cause of the change based on data associated with an optical performance parameter. The data is collected from optical network elements that have a value for the optical performance parameter exceeding a threshold. Another embodiment includes an electrical monitoring device, including outputting an electrical signal that indicates a degradation in the performance, and transmitting a command to the optical network elements to collect the data. A further embodiment includes an optical network element, including transmitting an optical signal to the electronic monitoring point that converts the optical signal to the electrical signal, which is examined to detect the degradation, and transmitting the data to the network management system, if the value exceeds the threshold value.