Learning Canonical Representations for Scene Graph to Image Generation
european conference on computer vision, pp. 210-227, 2019.
Results in Table 3 of the main paper demonstrate that the learned Soft Relation Closure model is more robust to changes in the scene graph input
Generating realistic images of complex visual scenes becomes very challenging when one wishes to control the structure of the generated images. Previous approaches showed that scenes with few entities can be controlled using scene graphs, but this approach struggles as the complexity of the graph (number of objects and edges) increases....More
PPT (Upload PPT)