New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Is Video Supported? #7

Open

ssmmoo1 opened this issue Sep 21, 2023 · 1 comment

ssmmoo1 commented Sep 21, 2023

Can this model accept video as an input?

Collaborator

gordonhu608 commented Sep 21, 2023 •

edited

Loading

Thank you for your interest in our work. Yes it also supported video input, see this line of code

BLIVA/bliva/models/bliva_vicuna7b.py

Lines 298 to 306 in b45425a

 if image.dim() == 5: 

 inputs_llm, atts_llm = [], [] 

 add_inputs_llm, add_atts_llm = [], [] 

 for j in range(image.size(2)): 

 this_frame = image[:,:,j,:,:] 

 with self.maybe_autocast(): 

 frame_embeds = self.ln_vision(self.visual_encoder(this_frame)) 

 frame_features =self.visual_encoder.get_intermediate_layers(this_frame)[-2]

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment