1 paper · 1 filter
Feng-Lin Liu, Hongbo Fu, Xintao Wang +4
Video generation and editing conditioned on text prompts or images have undergone significant advancements. However, challenges remain in accurately controlling global layout and g…