Pinned Loading

1. vinvl_bert vinvl_bert Public

   Vision-language model for generating Arabic image captions using Bidirectional Transformers (BiT) and advanced feature fusion.

   Jupyter Notebook

2. vinvl vinvl Public

   A simplified visual backbone for feature extraction, bounding boxes, and object detection using VinVL.

   Jupyter Notebook

3. violet violet Public

   Violet: A Vision-Language model for generating Arabic image captions using a Gemini Decoder and pretrained transformer.

   Jupyter Notebook

4. voca_grow_app voca_grow_app Public

   VocaGrow is an intelligent mobile application built with Flutter that empowers children to practice pronunciation and language skills through AI-driven voice interaction. The project is developed f…

   Dart 1

5. logo-similarity-retrieval logo-similarity-retrieval Public

   A deep metric learning pipeline for logo recognition and similarity search using synthetic data and unsupervised clustering.

   Jupyter Notebook

6. arabic-visual-question-answering arabic-visual-question-answering Public

   Modular and extensible framework for Arabic Visual Question Answering (VQA) using state-of-the-art pretrained models for image captioning and question answering.

   Jupyter Notebook