Abstract: This paper explores the effectiveness—specifically in improving video consistency—and the computational burden of Contrastive Language-Image Pre-Training (CLIP) embeddings in video ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback