OmniTokenizer: A Joint Image-Video Tokenizer for Visual Generation
Open paper
Benchmark
Model
Rank
Results
video-generation-on-ucf-101
OmniTokenizer-AR
#13
FVD16: 191
video-prediction-on-kinetics-600-12-frames
OmniTokenizer-AR
#8
FVD: 32.9
Rank counts only results with a code link.