1 paper · 1 filter
Mahdi Naseri, Jiayan Qiu, Zhou Wang
We study Generated Contents Enrichment (GCE), a conditional image-generation task in which a sparse scene description is first enriched through an explicit scene representation and…