**4. Experimental Details**

This section discusses the experimental details of our proposed CGANet model and the quality matrices used to evaluate the performance of the proposed model. CGANet performance is compared with two other state-of-the-art methods: the Gaussian mixture model [18] and lightweight pyramid networks [17]. The algorithm implementation was conducted using Python and TensorFlow 2.0 [46]. CGANet was trained on a computer with a 2.2 GHz, 6-core Intel core i7 processor, 16 GB memory, and an AMD Radeon Pro 555X GPU.
