Computer Vision
Collection
Computer vision models, datasets, etc.
•
9 items
•
Updated
A Persian image captioning model constructed from a ViT + RoBERTa architecture trained on flickr30k-fa (created by Sajjad Ayoubi). The encoder (ViT) was initialized from https://huggingface.co./google/vit-base-patch16-224 and the decoder (RoBERTa) was initialized from https://huggingface.co./HooshvareLab/roberta-fa-zwnj-base .
pip install hezar
from hezar.models import Model
model = Model.load("hezarai/vit-roberta-fa-image-captioning-flickr30k")
captions = model.predict("example_image.jpg")
print(captions)