With the exponential growth of visual applications, the perceived need for remote sensing images has been increased, which consequently raises the desire for image quality with more precise details. However, directly providing images that are full of spatial, spectral, and temporal information practically is not feasible. For this image, fusion has been found to be the best technique for achieving the desired results. The conventional fusion techniques are found to enhance the quality of the fused images through manually tweaking the algorithm rules; however, image fusion based on deep learning can continually adjust the quality of fused images through the training process, resulting in the expected outcome. This work attempts to summarize those deep learning techniques that have been reported on various remote sensing image datasets for improving image quality with expected details through image fusion. To demonstrate the efficacy of the suggested solution on the provided dataset for the observed findings, a thorough study of the most advanced deep learning-based image fusion solutions currently accessible has been presented. This work will be of interest to the working professionals who are working on various visual challenges in the application of remote sensing images.