Cross-domain Image Generation

Image composition—seamlessly inserting a user-provided object into a new visual context—is a long-standing challenge in computer vision and generative AI.…