Shortwave infrared (SWIR) polarimetric imaging has been found very effective in various applications. However, the low resolution of the SWIR camera severely limits the capacity of this technique. Image reconstruction methods have been developed to improve the spatial resolution, but these methods typically do not consider the polarized information that the images may contain. In this paper, we propose a high-resolution reconstruction method for SWIR images based on the spatial information of visible images without losing polarized information in the SWIR image. Experimental results demonstrate that this method is feasible to reconstruct high-resolution polarized SWIR images. We have also demonstrated its potential application in image fusion. (C) 2019 Optical Society of America