Vehicle Image Generation Going Well with The Surroundings

Since the generative adversarial network has made a breakthrough in the image generation problem, lots of researches on its applications have been studied such as image restoration, style transfer and image completion. However, there has been few research generating objects in uncontrolled real-world environments. In this paper, we propose a novel approach for vehicle image generation in real-world scenes. Using a subnetwork based on a precedent work of image completion, our model makes the shape of an object. Details of objects are trained by an additional colorization subnetwork, resulting in better quality of the generated objects. In addition, we suggest a problem of finding plausible locations of an object in real world scenes and propose a baseline solution for it. We evaluated our method by generating vehicle images using Berkeley Deep Drive (BDD) and Cityscape datasets, which are widely used for object detection and image segmentation problems. The adequacy of the generated images by the proposed method has also been evaluated using a widely utilized object detection algorithm.