Make Pixels Dance: High-Dynamic Video Generation

Benchmark Model Rank Results
text-to-video-generation-on-msr-vttPixelDance–FVD: 381CLIPSIM: 0.3125
text-to-video-generation-on-ucf-101PixelDance (Zero-shot, 256x256)–FVD16: 242.82
video-generation-on-ucf-101PixelDance (256x256, text-conditional)–FVD16: 242.82Inception Score: 42.10