Text2Video-Zero icon indicating copy to clipboard operation
Text2Video-Zero copied to clipboard

Cross Frame Attention vs Sparse-Causal Attention

Open HyeonHo99 opened this issue 1 year ago • 1 comments

Hi, your work is amazing!

After reading your paper, I have one question. What exactly is the difference between Cross Frame Attention and the Sparse-Causal Attention from the Tune-A-Video paper?

Thank you.

HyeonHo99 avatar May 14 '23 12:05 HyeonHo99

Hi @HyeonHo99,

we are attending only to the first frame, without updating any model weights. In fact, by attending to the first frame only, we obtain better results and at the same faster results compared to the SCA approach.

rob-hen avatar May 16 '23 23:05 rob-hen