5fa1a76
1
2
Encoder-decoder[[cv-encoder-decoder]] Vision models commonly use an encoder (also known as a backbone) to extract important image features before passing them to a Transformer decoder.
