Image fusion is utilized in remote sensing (RS) due to the limitation of the imaging sensor and the high cost of simultaneously acquiring high spatial and spectral resolution images. Optical RS imaging systems usually provide images of high spatial resolution but low spectral resolution and vice versa. Therefore, fusing those images to obtain a fused image having both high spectral and spatial resolution is desirable in many applications. This article proposes a fusion framework using an unsupervised convolutional neural network (CNN) and Stein’s unbiased risk estimate (SURE). We derive a new loss function for a CNN that incorporates a backprojection mean square error (MSE) with SURE to estimate the projected mse between the fused image and the ground truth. The main motivation is that training a CNN with this SURE loss function is unsupervised and avoids overfitting. Experimental results for two fusion examples, multispectral and hyperspectral (MS–HS) image fusion and multispectral and multispectral (MS–MS) image fusion, show that the proposed method yields high-quality fused images and outperforms the competitive methods. Codes are available at https://github.com/hvn2/Deep-SURE-Fusion .