Photo-realistic image stylization attempts to transfer style from one image onto a given image. Imagine going to London after a long semester, but the weather was cloudy. One would like to be able to take photos of the destination in any preferential scenario - sun, rain or snow. Another scenario would be for real-time image stylization i.e. filters. This has been a fairly well studied domain. The major challenge of this problem is to transfer the style onto a generated image that looks like it has been taken by a camera. Rendering the semantic content of an image in different stylizations seems to be an interesting and difficult computer vision task with a lot of applications, thus forming the basis of motivation for this project. Our aim in this project was to perform an evaluation of several image stylization techniques, and through our evaluation, determine where improvements could be made. Ultimately, we designed two approaches. The first improves upon the classic Whitening and Coloring (WCT) approach through the inclusion of a novel decoder transform, and then passes the resulting image through a spatial pooling network to reduce distortion. The resulting images are comparable to other approaches, and the approach demonstrates improvement in terms of speed and universality and avoids the need for segmentation maps or computationally costly post-processing steps. Our second approach was to design an automated, heuristic method of preparing segmentation maps for the existing WCT² approach, which relies on manually designed segmentation maps. The results of our approach compared favorably to those of WCT² without segmentation maps, and do not require a deep neural network or human guidance.