search

esdolo / FreeVS

FreeVS: Generative View Synthesis on Free Driving Trajectory
Apache License 2.0
61 stars 1 forks source link

which VAE model did you use？ #4

Closed SophieZhou closed 6 days ago

SophieZhou commented 1 week ago

The work is very nice and very sensible. But in the paper, you did not tell, which VAE model did you use. Does VAE effect the results? Is any VAE model all right?

SophieZhou commented 1 week ago

And by the way, how to generate the text messages or text annotations for each image?

SophieZhou commented 1 week ago

And by the way, how to generate the text messages or text annotations for each image?

In your paper, I have not noticed this information. How to describe the images, using "cars, color, street" all these information?

SophieZhou commented 1 week ago

And by the way, how to generate the text messages or text annotations for each image? I see, using images to generate text information by CLIP. Is it right.