Example‐based image recoloring in an indoor environment

Abstract

Color structure of a home scene image closely relates to the material properties of its local regions. Existing color migration methods typically fail to fully infer the correlation between the coloring of local home scene regions, leading to a local blur problem. In this paper, we propose a color migration framework for home scene images. It picks the coloring from a template image and transforms such coloring to a home scene image through a simple interaction. Our framework comprises three main parts. First, we carry out an interactive segmentation to divide an image into local regions and extract their corresponding colors. Second, we generate a matching color table by sampling the template image according to the color structure of the original home scene image. Finally, we transform colors from the matching color table to the target home scene image with the boundary transition maintained. Experimental results show that our method can effectively transform the coloring of a scene matching with the color composition of a given natural or interior scenery.

Deep Learning-Based Machine Color Emotion Generation

“…The objective assessment metrics used in this study were peak signal-to-noise ratio (PSNR) proposed by Wang et al (2004), structural similarity index (SSIM) presented by Wang et al (2003), and root mean square error (RMSE) shared by Deshpande et al (2015) and Larsson et al (2016). PSNR and SSIM are two measurement tools widely used for image quality assessment (Lin et al, 2020). PSNR is used to measure the difference between the generated color image and the ground truth image and evaluate the recovery algorithm's performance.…”

Animation line art colorization based on the optical flow method

“…Thasarathan et al 31 also used a model similar to cGAN for the colorization of grayscale maps or line sketches but without considering the consistency of the temporal dimension. Lin et al 32 proposed a framework for color migration of home scene images. After dividing the image into local regions and extracting their corresponding colors, the template image is sampled according to the color structure of the original home scene image to generate a matching color table, and finally, the colors are transformed from the matching color table to the target home scene image with the boundary transition maintained.…”