CVPR2020

Wish You Were Here: Context-Aware Human Generation

Oran Gafni, Lior Wolf

Abstract

Figure 1. The "wish you were here" application. Given an image with one or more persons (a), and an optional bounding box indicating where to add a new person (b), the method generates the pose of the new person (c). Then, given the appearance information of the target person (d), the method renders a new image (e) with that person.