Skip to content

Multiple Encoder Option for Training Base Codec #3

Description

@kjungwoo03

Thank you for your nice work!

I have a question about training this codec from scratch,

In your WACV'25 paper, the paper mentioned that input image x is encoded into both latent base (y_b) and top (y_t). For that process, mutiple_encoder option should be True in codec.

However, in training recipe you provided in README and also in pretrained codec's arguments, that option is set to be False.

Is that pretrained codec represents the method presented in your paper?

Thank you again for your nice work!

Best regards,

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions