Running
1
🖼➜🔤
Vision Encoder-Decoder for Image Captioning
What can lightweight models do?
🤖 VED for Brazilian Portuguese IC
A Comparative Evaluation of Transformer-Based Vision Encoder-Decoder Models for Brazilian Portuguese Image Captioning, by LAICSI (IFES).
Note Space to demonstrate the Vision Encoder-Decoder models in action
laicsiifes/swin-distilbertimbau
Image-to-Text • Updated • 389 • 2Note This Vision Encoder-Decoder (VED) is an union of Swin Transformer and DistilBERTimbau fine-tuned in Flickr30K Portuguese
laicsiifes/swin-gportuguese-2
Image-to-Text • Updated • 268 • 4Note This Vision Encoder-Decoder (VED) is an union of Swin Transformer and GPorTuguese-2 fine-tuned in Flickr30K Portuguese
laicsiifes/flickr30k-pt-br
Viewer • Updated • 31k • 169 • 2Note Flickr30K Portuguese Translation with Google Translator API