antoyang / FrozenBiLM

[NeurIPS 2022] Zero-Shot Video Question Answering via Frozen Bidirectional Language Models
https://arxiv.org/abs/2206.08155
Apache License 2.0
156 stars 24 forks source link

Do you L2 Normalize the CLIP Features? #16

Closed mmaaz60 closed 1 year ago

mmaaz60 commented 1 year ago

Hi @antoyang,

After extracting the features from CLIP-L/14, do you L2 normalize the features before passing it to the subsequent pipeline? Thanks

antoyang commented 1 year ago

Hi, the features are not normalized.