End-to-end Training for Text-to-Image Synthesis using Dual-Text Embeddings

DiffMRe-assign community